The purpose of this study was to find an effective way of identifying homogeneous family firm groups, based on the prevailing beliefs of the owning family on ownership, management and intergenerational transmission. We conducted a two-stage cluster analysis, using data from a representative sample of 240 Spanish family firms. The results showed three types of clearly differentiated family firms, each of them with a profile of different beliefs. Our work contributes to the previous literature by integrating the components of involvement and essence approaches and beliefs about these to distinguish homogeneous groups of family firms.
Despite the fact that family firms are one of the oldest types of business organisations and the important role these play in the development of economies around the world, there is not yet a consensus definition of what a family firm is (Cano-Rubio et al., 2017; Iturralde et al., 2011; Litz, 2008). The definitions used in previous literature have differed greatly, so much so that some authors have found up to 30 different ways of defining a “family firm” (Litz, 2008; O'Boyle et al., 2012). However, an in-depth analysis of the literature suggests that the majority of authors address the definition of family firm from two different but complementary theoretical approaches: (1) the involvement approach (Chua et al., 1999; Vallejo, 2007); and (2) the essence approach (Chua et al., 1999; Habbershon et al., 2003). In terms of the first approach, the distinction between family and non-family firms is based primarily on two elements: family participation in the ownership, and the family's participation in the management and control of the business. For the essence approach, one of the most important elements for defining a family firm is the transgenerational vision.
The main reasons that have been noted to explain the plethora of definitions of family firm are related to cultural differences, different legal frameworks or different definitions of a family that exist in different countries or contexts in which the definitions have been framed (Harms, 2014). However, whilst we can understand the reasons behind such a wide range of definitions, it seems illogical that, for a single sample of businesses, those classified as family firms should vary according to the different definitions used (Anderson & Reeb, 2003; Astrachan & Shanker, 2003; Klein, 2000; Sanguino, Barroso, & Bañegil, 2012). This shows the need to agree on an operational and functional definition that allows for clear distinctions between firms that are family firms and those that are not (Chrisman et al., 2010; Chua et al., 1999). That is to say, it seems necessary to reach a consensus on a definition that would make it possible to establish the study population with sufficient validity and reliability, which promote further study of these kinds of organisations and will allow results obtained in different scientific research projects to be compared.
On the other hand, in order to better understand and compare the results obtained in different national or international samples (Astrachan & Zellweger, 2008; Sacristán-Navarro et al., 2011), it must be remembered that there is great heterogeneity within the group of family firms (Westhead & Howorth, 2007). This diversity of family firms results in different behaviours in the various internal and external processes (e.g., the process of selecting a new leader) in which they are involved, and probably also leads to different performances that typically have little to do with the size, sector of activity or the country in which they are implemented.
In an attempt to find homogeneous groups of family firms, the previous literature has classified them by considering the evolution of the family firm throughout its existence (Gallo, 2004), the ownership structure (Lansberg et al., 1988), the interaction between the goals of the family and the business (Sharma, 2004), the control of ownership and the desire to keep the business in the family's hands (Vallejo, 2007) or its degree of professionalization (Camfield & Franco, 2019). According to Dekker et al. (2013), these classifications have their limitations and do not sufficiently reflect the wide diversity of family firms. As shown by Michiels et al. (2017), or Hernández-Linares et al. (2017), despite the fact that a classification system for family firms is increasingly necessary, researchers have not yet found a system that is operational and effective. Thus, based on the review of the different definitions and classifications of family firms, the main goal of our research was to develop a new way of identifying the different types of family firms. To this end, the research focused first on determining what elements are essential to identify a firm as a family firm. The research then analyses and reflects on the set of typologies and classification schemes of family firms that have been proposed in prior literature.
We propose that the heterogeneity of family firms should be approached using a classification that is modular and that takes the prevailing beliefs of the owning family on key aspects underlying the very definition of family firm into account, namely ownership, management and intergenerational transfer. Denison (1990) advocated that beliefs are at the heart of corporate culture. These beliefs lead to behavioural norms that guide the behaviour of individuals and groups within the organisation, and represent a source of competitive advantages that distinguishes some businesses from others. Thus, our research question was: “Can the predominant beliefs of the owning family about ownership, management and intergenerational transfer help to identify different groups of family firms?”
This paper makes at least two contributions to existing literature. First, we propose a system that is operational and effective for classifying family firms into homogeneous groups. We argue that classification should be based on the beliefs that the dominant group holds about the involvement of the family at the ownership/control level; the involvement of the family in the business's management; and the transgenerational vision. This multidimensional approach includes, for the first time in a taxonomy, the demand of some authors to integrate the main components of the essence and involvement approaches (Harms, 2014). Second, other research has adopted a narrower view, looking at the influence of the family on the business through the amount of family involvement in ownership and management (Dekker et al., 2013), and/or the desire to keep the business in the hands of the family through the generation they are in or the number of generations involved (Casillas et al., 2010; Kontinen & Ojala, 2010). We adopt a more holistic view, as we take into account the beliefs that the dominant group holds about both the essence approach and the involvement approaches. On the other hand, by modulating the beliefs in 3 different degrees, our classification offers a parsimonious and modular way of understanding the heterogeneity of family firms.
This paper is structured as follows: we first establish the theoretical background, in which we reflect on the factors that allow us to identify family firms and their heterogeneity. The paper then presents the sample and methodology, the main results and a discussion of them. Finally, we present the main conclusions, the limitations of our research and propose further developments.
2Theoretical background2.1Differentiating between family and non-family firmsIn the last 10 years, research related to family firms has grown exponentially (Araya-Castillo et al., 2021; Zellweger et al., 2010). A significant part of these studies has focused on explaining the differences between family and non-family firms, which has forced researchers to define what family firm means. A review of the literature shows a clear conclusion: there is still no consensus on the parameters that should be used to identify a business as a family firm (Harms, 2014; Iturralde et al., 2011; Litz, 2008).
The heterogeneity between the various definitions of what constitutes a family firm possibly derives from the importance that researchers and practitioners give to what they consider to be the main element or elements that differentiate a family firm from other types of businesses (Cano-Rubio et al., 2017; Vallejo, 2007). This heterogeneity could also be due to cultural or legal differences in the different countries or contexts in which these definitions have been framed (Harms, 2014). Some researchers have found that only one variable is needed to correctly identify family firms and they have proposed mono-criterion definitions. These include those that consider the most important element in defining a family firm as the control ownership of the capital (Davis & Harveston, 2000; Littunen & Hyrsky, 2000), the management (who makes decisions in the business) (Filbeck & Lee, 2000; Neubauer & Lank, 2003) or the concept of passing the business on to further generations (Sharma et al., 2001; Tan & Fock, 2001). However, other researchers have proposed multi-criteria definitions, which often combine two or more elements of the mono-criterion or even add other additional criteria (Fahed-Sreih & Djoundourian, 2006; Maseda et al., 2019; Shepherd & Zacharakis, 2000)
The involvement approach argues that the difference between a family firms and other organisations is in the involvement of the family in the ownership/control and management of the business. Therefore, these are the essential and sufficient components for identifying a business as a family firm (Pearson et al., 2008). This is the idea on which the multi-criteria definitions most frequently used by researchers are based, relying on a multidimensional view of family firms. From a practical point of view, a definition based solely on the levels of family involvement in the ownership and management of the business can lead to misclassification. For example, a business whose management is entirely in the hands of professionals outside the family but 100 % of the ownership corresponds to one family that controls the governing body and dictates the business's strategic decisions would be classified as a family firm. On the other hand, the involvement approach does not take into account two factors that are critical in explaining the different behaviours of family and non-family firms: altruistic behaviour and co-responsibility.
In contrast to the component of involvement approach, the essence approach suggests that identifying a family firm should focus on the behaviours that differentiate them from other businesses; behaviours that may not be captured when only taking into account the family's level of involvement in management and ownership/control. For the essence approach, these unique behaviours characteristic of family firms are due to the continuous interaction between the business and family systems (Habbershon et al., 2003). More specifically, they are due to the continuous interaction of the family's unique resources and capacities for the business and are perpetuated thanks to the transgenerational business vision (Suess-Reyes, 2017). From this point of view, the main element of the definition of family firm is the transgenerational business vision (Westhead & Howorth, 2007). This transgenerational vision influences the behaviour of those who control the business and helps to perpetuate the resources and capacities that differentiate it.
For Chrisman et al. (2005) or Harms (2014), the main elements of the essence and involvement approaches should be considered complementary and not antagonistic. As shown by Harms (2014), “current research came to the agreement that both concepts should be integrated in a well-grounded family firm definition”. According to this criterion, in order to identify family firms and differentiate them from other businesses there are three main elements to take into account: the involvement of the family at the ownership/control level, the involvement of the family in the business's management and the transgenerational business vision. This multidimensional and inclusive vision results in an operational concept of family firm, which provides a quick, clear and systematic way of identifying family firms. In addition, it is a “universal” concept that can be applied irrespectively of the definition of family firm that we use, the cultural or legal framework in which it applies, and the business and/or family's stage of development.
2.2Heterogeneity of family firmsIn all family firms there is a distinct element in common, namely the coexistence of two systems that overlap and evolve interrelatedly: the business and the family. However, not all family firms are equal and it would be a mistake to consider them together as a single group. To understand the behaviour and results of family firms we need to adequately differentiate between different types of family firms. That is to say, we cannot consider family firms to all belong to a single model (Dekker et al., 2013) because in reality there are many different kinds of family firms.
The previous literature has proposed different ways of identifying specific groups of family firms (Astrachan et al., 2002; Dekker et al., 2013; Harms, 2014; Sharma & Nordqvist, 2008; Vallejo, 2007; Westhead & Howorth, 2007). With some exceptions, most of the classifications are based on the varying amounts of family involvement in ownership and management. These classifications assume that the degree of family involvement is sufficient to explain all the behaviours of the family firms and therefore their outcomes. According to Dekker et al. (2013), this assumption is erroneous and such classifications are unable to differentiate between family firms that behave differently. The components of the involvement approach are easily measurable but do not seem sufficient in explaining the different behaviours and strategic processes that lead to the competitive advantages of different types of family firms. In addition, the components of the involvement approach are necessary, but are insufficient in identifying the different family firm groups. We believe that this classification should also consider the main element of the essence approach: the transgenerational business vision. Depending on the nature of the transgenerational business vision, the altruistic and co-responsible behaviours of those who control and manage the business may be different and influence the way in which it creates competitive advantages differently (Westhead & Howorth, 2007). Taking the transgenerational vision into account means reflecting the family's idiosyncrasies and influence on the business's processes and resource base in the classification. It also means considering the behaviours that differentiate some family firms from others and the transfer between generations.
In addition to the essential components that appear in the definition of family firm proposed in the previous section, based on integrating the involvement and essence approaches, other classifications propose considering the culture (Astrachan et al., 2002), the identity of the family firm (Zellweger et al., 2010), or its values (Rau et al., 2019). In line with these suggestions, we propose that the classification of family firms take into account whether or not the family forms a substantive part of the business or whether, on the contrary, the family is only a symbolic element that does not influence the business's behaviour and is not integrated into the organisation's culture. More specifically, we propose that the classification covers the family's beliefs on control of ownership, control of management and intergenerational transfer. As argued by Scott and Lane (2000), beliefs are key to differentiating some organisations from others and tend to be unwavering and indisputable, given their deep roots due to the validation mechanism in the environment (Castresana & Blanco, 2002). Beliefs are the principles or convictions that people consider to be true and are willing to defend. Beliefs are the point of reference for understanding the decisions and behaviour of family firms. They play a very important role in their structure, culture and strategy (Rau et al., 2019). On the other hand, previous research has shown that beliefs are related to the performance of the family firms, as they are a source of sustainable competitive advantage (Habbershon et al., 2003), and contribute to a long-term vision (Sharma & Nordqvist, 2008). Leung et al. (2002) demonstrated that there is a close and significant correlation between beliefs and the behaviour of managers.
Beliefs are transmitted in family firms through what is said and what is done, through education and socialisation processes that lead to its members forming an opinion which is believed to be true and unchanging at that time (Gehman et al., 2013; Jaskiewicz et al., 2016). This means that the transmission of different beliefs amongst members of family firms could result in family firms with different structures, cultures, strategies and performance (Brice & Jones, 2008). Davis (1984) defended the relationship between culture and strategy, arguing that strategy responds to the beliefs, the reasons behind the organisation wanting to comply with that strategy. The internalisation of these beliefs by members of the family firm leads to higher performance (Dyer, 2006). Therefore, it seems reasonable to assume that the beliefs of the owning family offer a justification for the heterogeneity observed within the group of family firms. Thus, the classification that we propose recognises that family firms develop resources and capabilities that differentiate them from other family firms: (1) through the continuous overlap and interrelation of the business and family systems (Barnett et al., 2009; Sundaramurthy & Kreiner, 2008); (2) through the different degrees of involvement and influence that the family has on the business (Chrisman et al., 2012); and (3) through the process of transmission of the family identity, and the behaviours that characterise this, between generations. However, what really differentiates family firms are the whys, the beliefs, the acceptance of truth on how the ownership, management, and intergenerational transfer should be. The family group's differing beliefs about the percentage of ownership that should remain in the hands of the family and under what circumstances outside partners should be allowed to enter; beliefs about the circumstances under which outside managers should or should not be allowed to be included in the management of the company; and/or beliefs about the prioritisation of individual desires and rights over those of the family as a whole in generational transmission, can have a significant influence on the degree of family involvement in the company (Rau et al., 2019), on its financial structure (Acedo-Ramirez et al., 2017) or on its success (Manzano-García & Ayala-Calvo, 2020). It is the integration of various beliefs that originate in the family and the business, often linked to a common history, that modulates the different degree of involvement and influence that the family exerts on the business. It is this integration that ultimately allows us to explain the differentiating characteristics of the resources and capacities that one group of family firms possesses versus another. However, at present, to our knowledge, there are no studies that have used the beliefs of the owning family to classify family firms into homogeneous groups.
3Sample, measurements and methodology3.1Sample and procedureIn this study, a business will be classed as a family firm if it meets three conditions: (1) a substantial portion of the shares are held by the founder or family members, allowing them to exercise control over the business. Like Acedo-Ramirez et al. (2017) and Schepers et al. (2014), we established 50 % as the minimum percentage of a business's equity that should be in the hands of the founder or family members in order to consider the business to be controlled by a family; (2) the family actively participates in monitoring the business. In accordance with the most used definition in empirical studies (Chen et al., 2010), we consider that the participation of at least one family member in management positions (either on the Board of Directors or in management) means active participation in the monitoring of the business; (3) the business is in its second or even later generation. Some authors consider it sufficient that there be an intention to transfer the business to the next generation to classify a business as having a transgenerational vision (Kontinen & Ojala, 2010; Sanguino, Barroso, & Bañegil, 2012). We use a more restrictive criterion in accordance with the authors who consider that a family business demonstrates its transgenerational vision when it has been transferred to the members of the next generation at least once (Rau et al., 2019).
A telephone questionnaire survey research method was used in this study. The interviewers called 815 CEOs of companies that met the three criteria for inclusion in the sample described above. The CEOs were randomly chosen from the Family Business Institute database (Corona, 2018). Of the 253 CEOs who agreed to participate in this study, a total of 240 valid surveys were obtained (overall response rate of 29.4 %). All data were collected in the first quarter of 2019. In accordance with the European Union criteria, 87.31 % of the sample businesses are small and 12.69 % are medium-sized (Commission of the European Communities, 2003). These data show that the characteristics of our sample are very similar to those shown by family firms in Spain Corona (2018).
3.2Questionary designThe review of previous literature revealed that there are no existing scales for measuring the dominant beliefs on ownership, management and intergenerational transfer. For this reason, we developed our own measurement scales. The draft questionnaire was independently reviewed by researchers with ample experience in the family firms culture. Next, we wrote a pretest taken by 20 CEOs of family firms located in the Autonomous Community of La Rioja. This pretest allowed us to check the internal consistency of the constructs and find out if all the items were understood correctly. The results of the pretest and a preliminary factor analysis suggested that some items be eliminated (items with factor loadings above 0.5) and others be re-worded. Based on the results of this analysis and having checked the internal consistency of the scales and their convergent and divergent validity, the questionnaire was adapted with minor changes.
3.3MeasurementsSocio-demographic variables. Each respondent was asked about the percentage of the business's equity in the hands of the founder or family members, the number of family members in management positions and which generation the family firm was in, to verify that all businesses in the sample could be classed as family firms and met the inclusion criteria. The data relating to the size of the business (number of employees, turnover and total assets) were taken directly from the SABI Database (Bureau van Dijk, 2018), a database that contains economic and financial data on over 1250,000 Spanish businesses.
Beliefs. We used three items to collect information about beliefs in relation to the business's ownership (BO). Each CEO had to choose one of the following options: (1) “Ownership must include partners outside of the family that are necessary in order to properly develop the business's activity”; (2) “Ownership can include external partners but the majority of the shareholders' meeting should remain in the hands of the family” and (3) “Complete ownership must remain within the family”.
We used three items to collect information on beliefs about who should manage the business (BM). Each CEO could choose from the following options: (1) “The business's management must be in the hands of the best professionals possible within our field of business, even if they are not members of the family”; (2) “The business's management must include external professionals but should give priority to members of the family, especially in key positions” and (3) “The business's management must be in the hands of one or more members of the family, only resorting to external professionals when there are no candidates from within the family”. In terms of beliefs on intergenerational transfers (BIT), respondents were asked to choose one of the following options: (1) “In the transfer of the family business, the wishes and rights of the family's members should be prioritised”; (2) “In the transfer of the business between generations, the business must be seen as a collective heritage that is managed according to the wishes of the majority of the family” and (3) “In the transfer between generations, the business should be considered as a legacy for future generations of the family, members of the family can enjoy the usufruct but cannot sell the common heritage".
The detailed operationalization of all the variables can be found in Table 1.
List of variables, items and scales.
In order to test whether it is useful to distinguish beliefs regarding ownership, management and intergenerational transfer amongst family firms, a Cluster Analysis of cases was carried out. As suggested by Hair et al. (2014) or Rau et al. (2019), “the cluster analysis is an ideal approach to classify observations into similar sets or groups and develop a taxonomy”. More specifically, following Punj and Stewart (1983) and Hair et al. (2014), we performed a two-stage cluster. In the first stage we performed a hierarchical cluster analysis, which allowed us to determine the number of groups and the value of their centroids. We generated a hierarchical dendrogram, a visual representation of the steps in hierarchical cluster analysis, and an agglomeration schedule table that showed the combined clusters and the values of the coefficients at each step. A large percentage change in the agglomeration coefficient indicates that two nonhomogeneous groups will be combined in the further agglomeration (Hair et al., 2014). In the second stage, starting from the non-random solution generated in the first stage, we conducted a K-means cluster analysis. We used the K-media reassignment method that allowed us to separate cases into K clusters, so that each case belonged to a single group. In order to determine the optimal number of clusters to be considered, and to maximise homogeneity within the groups, a range of solutions was tested. The results of each solution were filtered using univariate analysis of variance (ANOVA) and discriminant analysis (Luque, 2012). Ideally, the mean value of the dependant variables (BO, BM and BIT) should be significantly different in each of the groups or clusters. On the other hand, based on the discriminant analysis, the best solution is one that simultaneously generates the highest goodness-of-fit (square of significant canonical correlations) and the highest percentage of well-predicted cases (predictive ability).
Although some authors argue that hierarchical cluster analysis is the most appropriate when the sample size is small (Hair et al., 2014), two-stage cluster analysis has two advantages to consider: (1) it avoids the drawback of hierarchical procedures: since each unit of analysis is classified only once, there is a risk that the initial union is not actually suitable. This possible error is solved in the second stage, when performing the K-mean cluster analysis; (2) in the K-mean cluster analysis, starting from a “refined” initial solution (the non-random solution generated in the first stage) prevents the final solution from being a local optimum but not the classification optimum (Hair et al., 2014).
4ResultsIn our study, the percentage change in the clustering coefficient increased considerably when the number of clusters was reduced from three to two, indicating that three clusters would be sufficient to describe the sample. Thus, we conducted K-means cluster analysis to generate the three clusters. To ensure stability of the results, we iterated by generating from two to five clusters.
Table 2 shows the results of the ANOVA analyses and the discriminant analysis for the range of cluster solutions proposed. In our case, the three-cluster model was selected as the final solution. This decision was based on the number of significant ANOVAS and the number of significant discriminant functions and their explanatory and predictive capacity.
Table 3 shows, for each cluster, the number of companies classified in this cluster (cluster size) and the average value of each of the three belief categories considered in the analysis (BO, BM and BIT). To measure the quality of the clustering solution we used the entropy criterion (McLachlan & David, 2000). The entropy score I (3) = 0.95, which indicates a very good separation between the three clusters.
Based on the results shown in Table 3, we can define the three clusters as follows:
Cluster 1 – Includes a total of 32.5 % of the sample businesses. The dominant group of these family firms believe that ownership must remain in the hands of the family (F1). This means that they are willing to renounce an increase in the number of partners, in order to have a decisive influence in defining the business's mission and vision, as well as in its growth strategies. In short, they are family firms in which the family fully controls the business plans. Regarding the business's management, they believe that it should be in the hands of one or more members of the family, only resorting to external professionals when candidates within the family are not found (F2). In other words, they believe that the majority of executive positions should be occupied by family members. These family firms have a very small level of openness to outsiders, which means greater family involvement at all levels of the organisation. In terms of intergenerational transfer, they believe that the business should be considered as a legacy for future generations of the family. The members of the family can enjoy the usufruct but cannot sell the common heritage (F3). This means that the current owners believe that the business belongs to all members of the family and that it should remain so in the future. The owners feel obliged to keep the business and its culture to transfer these to the next generation. In addition, this entails the need to jointly and severally assist family members that have financial needs, as they will not be able to sell their part of the business to resolve these needs. For family firms in this cluster, the family is a substantive part of the business. As such, it is likely to observe altruistic, loyal, co-responsible and trusting behaviour in these types of family firms. We have labelled this cluster “Closed family firms” (the family comes first).
Cluster 2 - This cluster covers 26.67 % of the family firms in the sample; it is the cluster with the fewest businesses. It is characterised by the fact that, on average, the beliefs that prevail in the family firms are: (a) that the ownership must incorporate partners outside of the family, in the amount that may be necessary for the proper development of business (F1). These family firms are not concerned with safeguarding family ownership, control and financial independence. On many occasions, the desire to maintain family control of the business impedes the entry of new partners and access to other sources of external funding, which is one of the main causes affecting the growth opportunities of family firms (Romano et al., 2001). This does not occur in this cluster's family firms. These are family firms that prioritise the development of business over its control. Control of the general meeting of shareholders or the board of directors does not matter as much as the business's expansion; (b) management must be in the hands of the best professionals, whether or not they are members of the family, so that the best professionals are always leading the different processes involved in managing a business (F2). These family firms are willing to open their organisation to external managers that often coexist with family managers, creating a hybrid management system (Zhang & Ma, 2009). They are family firms with a high degree of professionalism, in which family involvement in day-to-day operations is usually small and authority has been decentralised; (c) the wishes and rights of the family members must be prioritised in the intergenerational transfer (F3). This means that the business's ownership will be transferred to the next generation in the form of individual shares or units, and each member of the family will be able to sell them when needed or they deem appropriate. Unlike cluster 1, the dominant group of family firms in cluster 2 believes that involving new partners can be positive for the business. They also do not share the idea that 100% of the business must remain in the hands of the family generation after generation. In summary, family involvement is much lower for the businesses in this cluster than those in cluster 1 and the businesses are much more open to including non-family members in both the business's ownership and management. We have labelled this cluster as: “Opened family firms” (the business comes first).
Cluster 3 - This cluster is formed by the largest number of family firms, 40.83% of the sample. It is characterised by the fact that, on average, they believe that: (a) the ownership must incorporate partners outside of the family, in the amount that may be necessary for the proper development of business (F1). In terms of management, they believe that management must include external professionals but priority should be given to members of the family, especially in key positions (F2). Regarding the intergenerational transfer of the business, they believe that this must be seen as a collective heritage that is managed according to the wishes of the majority of the family (F3). If we could place family firms on a continuum that ranges from “closed family firms” to “opened family firms”, we could say that the family firms in this cluster are in the middle of the two. These are family firms that believe in exercising control of strategic decisions not through a firm grip of ownership but rather through the leaders that make these decisions. Regarding transfer, the dominant group's beliefs are that the wishes of the majority of the family should be respected, which places it between those who state that individual rights must prevail and those who believe that the business is a legacy to be preserved in the hands of these family members. We have labelled this cluster "balanced family firms" to indicate that these are firms that might share some beliefs with cluster 1 or cluster 2, but that have other beliefs that are located between the two sets of beliefs displayed by these clusters. That is to say, as the beliefs go from “family first”, which indicates an almost complete overlap between the business and family's culture, to “business first”, which indicates minimal overlap between the two cultures, we find different types of family firms, ranging from closed family firms to opened family firms.
The three family firms types derived from the cluster analysis are represented in Fig. 1, with respect to the three beliefs categories.
5Discussion
This study, in response to our research question, contributes to the field of family firms research in several ways: First, it offers a model for differentiating some family firms from others. It is a multidimensional and inclusive model. It is multidimensional because it covers three essential factors in defining family firms; the two essential components of the involvement approach and the essential dimension of the essence approach (Harms, 2014; Suess-Reyes, 2017). Second, it is an inclusive model because it recognises that each of these three elements can interact with the others, creating a variety of resources and capabilities that are unique and inimitable for each family firm (Maseda et al., 2019). In addition, this is a novel model as it is based on the idea that beliefs on ownership, management and transgenerational transfer change the degree of involvement and influence that a family exerts.
Based on the results of the cluster analysis we have identified three types of clearly differentiated family firms: closed family firms, balanced family firms and opened family firms. This result supports previous studies that argue that family firms are a heterogeneous group within which different groups can be defined (Harms, 2014; Vallejo, 2007). However, our approach to the problem is different. Our classification is not based on the amount of family involvement in ownership and management, the number of generations who work in the business and/or the desire to keep the business in the hands of the family (Cano-Rubio et al., 2017). Our classification is based on the beliefs of the dominant group on the three key factors involved in defining a family firm. To some extent, our arguments match those of Astrachan et al. (2002), who propose that one must consider the overlap between the values of the family and the business and the family commitment to explain the family's influence on the business. However, our approach differs from the classification of family firms proposed by Astrachan et al. (2002) in two ways. Firstly, two subscales of their approach are based, respectively, on the amount of family involvement in ownership and management and on the number of generations involved in the business's management and, secondly, the overlap between the family and business's values and the family firms commitment are two components of a third subscale (culture), which is considered to be independent from the other two scales.
Our proposal is also close to the approach of Zellweger et al. (2010) or Rau et al. (2019). The former argue that the focus should be placed on the family firm identity that covers how the family defines and sees the business; the latter argue that the heterogeneity of family firms is due to the heterogeneity of the values integrated in their culture. One of the main differences between our work and that of Zellweger et al. (2010) is that we go a step further and propose a tool that, in addition to integrating the different theoretical approaches, allows family firms to be classified in different groups. The work of Rau et al. (2019) uses a set of 6 values to identify homogeneous groups of family firms. The most obvious difference with our work is that we link the beliefs of the dominant group with each of the three essential dimensions for defining family firms (ownership, management and intergenerational transfer); the work of Rau et al. (2019) does not provide these links.
Considering the size of the companies in our sample, where 87.31 % are small business according to the definition of the European Commission, it would seem logical to think that all of them have a moderate growth strategy over the generations. However, the modulation of this strategy could be influenced by the dominant beliefs in the family firm. Closed family firms, when they need to grow, are reluctant to resort to debt due to the subsequent loss of control, rejecting profitable investment projects because they believe that control of the company should remain 100% in the hands of the family (Acedo-Ramirez et al., 2017; Hernández-Trasobares & Galve-Górriz, 2016). On the other hand, one of the fundamental objectives of small family firms is often long-term survival (Cano-Rubio et al., 2017; Poutziouris et al., 2006), which leads successors to take fewer risks than their parents. In closed family firms, compared to opened family firms, successors have a stronger belief that wealth preservation is more important than wealth creation (Kaye & Hamilton, 2004); they believe that the business belongs to all members of the family and that it should remain that way in the future. These reasons, amongst others, will lead closed family firms to practice a less ambitious growth strategy than opened family firms, which in order to grow will be willing to take on debt or even include non-family partners in the shareholding.
6ConclusionsOur results showed that it is possible to find homogeneous groups of family firms by taking into account the beliefs of the dominant family group about ownership, management and intergenerational business transfer. These results contribute to covering several gaps identified in the previous literature. On the one hand, for the first time in a taxonomy, the components of involvement and essence approaches are used together in order to identify homogeneous groups of family firms. We adopt a holistic view, as we take into account the beliefs that the dominant group holds about both the essence approach and the involvement approaches. On the other hand, we propose a system that is operational and effective for classifying family firms into homogeneous groups. This classification offers a parsimonious and modular way of delimiting the heterogeneity of family firms; represents a further step toward and understanding the diversity of family firms. In addition, this study contributes to the consolidation of research on family firms, particularly regarding the definition of the object of research, and the consolidation of the results obtained. In other words, identifying the different groups of family firms could contribute to ensuring that the samples used in the research do not 'mix oranges with apples'(Dyer, 2006); therefore, contributing to the decline of the ambiguous and/or contradictory results that previous research has yielded (Harms, 2014).
Besides contributing to a better understanding of the differences between family firms, this work has important implications for researchers, practitioners, and even for teachers. Research on family firms has grown exponentially in recent years. However, there are still only a few studies that consider that the group of family firms is not homogeneous. This may be one of the reasons why investigations have yielded conflicting results. Another explanation of this fact may be that the tools used to identify the various types of family firms are very different. The methodology proposed in this study may serve as a starting point for future research to better understand the relationship between the different types of family firms and their financial performance, their innovation capacity, or their ability to manage conflicts, success in the probate process, etc.
We believe that there are several implications for practitioners. Knowing the key beliefs of the family firm with which you are working can help consultants to correctly design the family firm's governing bodies, to mediate business/family conflicts with greater success and to better advise in the drafting of family protocol, in the management of the probate process, etc. In addition, the information on the group to which the business belongs can also help credit institutions as beliefs about ownership, management and the transmission of the business provide information about the importance given to loyalty, reputation, who will manage the business in the future, or even on the collateral that can be made available to the lender; all of these variables can influence the amount and the cost of credit granted.
In addition, if teachers understand that not all family firms are equal, that their behaviours are different and that the same process (e.g., choosing a new leader) takes place differently depending on the group to which the business belongs, they will be able to design their programmes taking these differences into account. That is to say, both university and executive courses should be taught in a way that students can understand the differences between family firms and how the different beliefs on ownership, management, and intergenerational transfer lead to different behaviours and create different resources and capabilities.
This study, along with most others, is not without limitations that, in turn, offer new research opportunities. Firstly, all family firms in the sample are Spanish. Although our classification is not influenced by the definition of a family or by the legal context in which the sample is taken, it might be interesting to validate our results with samples from other countries. On the other hand, the family firms involved in this study were classified in small and medium companies. Further characteristics were not indicated such as business sectors, second, third or later generation or educational background of the CEOs. This summarises that the findings of this study are a really great starting point to the research community to go deeper into this triangle of ownership, management and transgenerational transfer. Secondly, as far as we know, no previous empirical research has used the scale for measuring beliefs about ownership, management and intergenerational transfer to classify family firms. The scale of measurement used only has three cut-off points. In the future, this scale could be reviewed and expanded to more accurately cover the different beliefs about each of the key elements of the definition of family firm. Thirdly, the profile of the three types of family firms found with the cluster analysis could be expanded by considering other quantitative or qualitative variables. Accordingly, our work does not investigate the relationship between the three types of family firms and the possible financial and non-financial outcomes. Future research should determine whether the propensity for innovation, propensity for intra-entrepreneurship, more or less ambitious growth strategies, financial results, etc. are more applicable to one type of family firms than another. The research findings also provide the basis to combine the identified family firm type with other research strands like knowledge sharing, knowledge transfers during succession phases. Finally, we collected data from CEOs to determine the beliefs of the dominant group. It would be interesting to collect information from other influential members of the family, both from those who work as well as those who do not work in the business.
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
