Internal Versus External Ownership Transition in Family Firms: An Embeddedness Perspective

39

22








In Swedish:"flergenerationsregistet"
INTERNAL VERSUS EXTERNAL OWNERSHIP TRANSITION IN FAMILY FIRMS:
AN EMBEDDEDNESS PERSPECTIVE
Johan Wiklund
Whitman School of Management
Syracuse University
721 University Avenue
Syracuse, NY 13244-2450
E-mail: [email protected]
and
Jönköping International Business School, Sweden

Mattias Nordqvist
Jönköping International Business School
PO Box 1026
SE-551 11 Jönköping
Sweden
Tel: +46-708-825624
E-mail: [email protected]

Karin Hellerstedt
Jönköping International Business School
PO Box 1026
SE-551 11 Jönköping
Sweden
E-mail: [email protected]

Miriam Bird
Stockholm School of Economics
PO Box 6501
SE-113 83 Stockholm
Sweden
E-mail: [email protected]


Keywords: succession, ownership transition, family firms


INTERNAL VERSUS EXTERNAL OWNERSHIP TRANSITION IN FAMILY FIRMS:
AN EMBEDDEDNESS PERSPECTIVE

ABSTRACT
We investigate factors that influence family business owners' choice between passing ownership within the family or to new external owners. Taking an embeddedness perspective focusing on owner-family structure and involvement we hypothesize that ownership dispersion, number of potential heirs, multigenerational involvement and whether the CEO is a family member, influence the choice of an internal or external transition of ownership. We build a longitudinal dataset from a sample of 3,829 family firms and their ownership transitions. Our theorizing and findings regarding ownership transitions complements the abundant research on management succession and therefore constitutes an important contribution to the literature.
INTRODUCTION
Succession continues to be the most extensively researched topic within the family business literature (see Yu, Lumpkin, Sorenson & Brigham, 2012). To the extent that ownership transition is considered (as opposed to management transition), the succession literature assumes that transition of ownership to the next generation is the preferred choice. However, as an alternative to passing on the ownership of their businesses to the next generation, owners can choose to exit ownership and sell their firms to outside parties if they regard this as a more attractive option). Rather than viewing external ownership transition as a last resort, we posit that the choice between internal and external ownership transfers among family firms depends on the owner-family's structure and involvement. Our argument is that these circumstances have an impact on the family's likely commitment to the business as well as family members' interest in keeping business ownership within the family.
Our research provides important contributions to the entrepreneurship and family business literatures. First, we focus on ownership transition rather than management succession. While the drivers and consequences of appointing a new chief executive (CEO) is a recurring theme in the family business succession literature, factors driving the choice between internal or external transition of ownership and the consequences thereof have not received much attention in the literature (cf. Ucbasaran, Wright & Westhead, 2001; Long & Chrisman, forthcoming). Given the extensive literature on succession, the paucity of research on the determinants of internal versus external ownership transitions is highly surprising for two main reasons.
First, existing empirical evidence suggests that external ownership transition is quite common. Although numbers are uncertain and vary across studies, it seems that only 20–30% of all family businesses are transferred to the next family generation (e.g., Sardeshmukh & Corbett, 2011). Such observations seem to run counter to the common assumption that if an heir is available, owners will prefer inside transfer of ownership (De Massis et al., 2008), suggesting that there are substantial gaps in our understanding of transition of ownership in family firms.
Second, family firms that go through an external ownership transition continue their existence, but they are no longer family firms (Ucbasaran et al., 2001). Such ex-family firms have received scant attention from scholars, who tend to see the sale of a family firm as failure regardless of the impact it has on the family, its financial outcomes, and the firm's post-succession performance (Ward, 1987, Zellweger, Nason & Nordqvist, 2012). Nevertheless, given that so many firms transition from the family business category, the choice of transition mode is worthy of more conceptual and empirical attention. It seems that to date, substantially much more attention has been paid to the minority that actually does transition to another generation of family members than the majority that does not.
Our second contribution relates to the understanding of the interplay between the business and the family. While most family business research still utilizes theoretical perspectives from the business side, we suggest that the family exerts a substantial influence on the family firm (cf. James, Jennings & Breitkreuz, 2012). Drawing on the embeddedness literature (Granovetter, 1985; Moody & White, 2003), we posit that the business is embedded in the family and that family and business are intertwined. Therefore, the choice between internal and external ownership transitions is influenced by owner-family structure, family involvement, and relationships within the owner-family. Our theorizing and findings regarding these factors provide a valuable counterweight to research examining business-related influences on family business transitions.
Third, we rely on variables related to the structure of the family and family members' involvement in the business (e.g., ownership dispersion, the presence of young and adult family members who do not hold ownership positions, intergenerational ownership, family CEO) as indicators of family structural cohesiveness and the embeddedness of the business in the family (cf. Moody & White 2003). These variables represent tangible aspects of the family. Access to such information should be reasonably available compared to perceptive measures of structural cohesion and/or embeddedness. For example, in this research, we were able to rely on secondary data for gleaning this information.
Our final contribution relates to the validity of our findings. It is difficult to obtain representative samples of family firms, and even if a sample is representative, low response rates typically threaten representativeness. Consequently, making statistical inference from a sample of family firms to a population can be challenging. Our results are based on examining the whole population of family firms (with 10 or more employees) within a country. As such, the findings represent the true image of family firms in Sweden.

THEORY AND HYPOTHESIS DEVELOPMENT

In line with recent suggestions to expand family business research by using social and family theories to understand strategic behaviors (James et al., 2012), we draw on the theories of embeddedness and structural cohesion to develop our predictions with regards to how family structure, social ties, and involvement impact the choice between internal and external ownership transitions. As James et al. (2012) observed, to date, family business research has not paid sufficient attention to the reciprocal relationship between the business and the family. In this context, an embeddedness perspective addresses how the business is embedded in the family and how family and business are intertwined (Aldrich & Cliff, 2003).
The embeddedness approach emphasizes social norms of exchange (characterized by closeness and trust) rather than formal contractual norms (characterized by calculation and opportunism) (Uzzi, 1997). Individuals' economic actions are embedded in social structures consisting of ongoing networks of interpersonal relationships and social ties that both facilitate and constrain actions (Granovetter, 1985). Building on this logic, Aldrich and Cliff (2003) introduced a family embeddedness perspective arguing that as an important social institution, the family has the potential to exert a substantial influence on the firm. The strength of family ties influence actors' choices and decisions because ties of different strength provide access to different kinds of resources (Granovetter, 1985). Importantly, actors and businesses differ with regards to how embedded they are in certain relationships and ties.
Also building on the embeddedness perspective, Moody and White (2003) used the notion of structural cohesion to show the circumstances under which a group's social relationships and hence cohesion may weaken. In essence, structural cohesion emphasizes the relational component of solidarity; that is, social relationships among group members are especially important because they are the forces and bonds that hold the group together. Thus, groups rely on social relationships between individuals to retain their character as a group. Further, cohesive groups feel a sense of togetherness and social solidarity towards each other (Moody & White, 2003). Any social group (e.g. an owner-family) can be characterized by such cohesiveness and solidarity. However, groups possess varying levels of cohesion and connectedness depending on the strength and structure of relations as well as the involvement of certain individuals.
From this perspective, an important task for family firm research is to study under what conditions a family as a social group retains its specific characteristics (James et al., 2012). In our case, this means whether the family retains its ownership of a business. Scholars typically describe the family as a structure constituted by strong ties, strong cohesion, and solidarity between its members. However, due to the changing nature of families, there is variance in the extent to which the business is embedded within the family. Hence, the strength of the relationships among family members may vary (Aldrich & Cliff, 2003). Furthermore, family firms are not homogenous in terms of family involvement and the family's commitment to the business (Melin & Nordqvist, 2007).
We argue that the degree of togetherness and connectedness related to the structural cohesion among embedded actors is important for understanding the choice between internal and external ownership transition..We expect that an owner-family characterized by weak togetherness and connectedness is more likely to lack solidarity and is more likely to choose an external ownership transition and vice versa. Pursuing this central thesis, we assume that the number of potential heirs, family ownership structure, intergenerational ownership involvement, and the appointment of an internal family CEO are relevant indicators of an owner-family's structural cohesion and thus their inclination to choose either internal or external ownership transition. We now develop hypotheses for each of these dimensions.
Family Ownership Dispersion
Ownership in family businesses has often been assumed to be concentrated, leading to a power structure in which decision-making authority rests in the hands of a single person or a few individuals who are closely tied together through family bonds (Carney, 2005; Jensen & Meckling, 1976). However, as the owner-family grows, ownership becomes dispersed among a wider set of individuals (Gersick et al., 1997; Schulze et al., 2003). While increased family ownership dispersion may have positive implications, such as a larger pool of competencies, more diversity in viewpoints upon which to draw, and potentially greater access to human and financial capital, it is also likely to be associated with greater potential for weak cohesion and a lack of feeling of togetherness and solidarity among owner-family members (Kellermanns & Eddleston, 2004; Schulze et al., 2003). If the group of owners is small, group members are more likely to have direct connections to each other, leading to higher network density within the family firm (Moody & White, 2003). This high network density facilitates the emergence of strong ties and leads to increased cohesion among the owner-family. As the group of family owners is extended the probability that actors are directly linked to all other members of the group decreases. This leads to weaker relationships. In other words, more owners is likely associated with weaker social relationships and cohesion. Hence, it is less likely that all owners identify themselves with a shared strategic agenda. Diminishing social cohesion may be also result from disagreements regarding sharing of dividends between owners who are operatively involved in the business and those who are only shareholders or who should be offered employment in the business (Eddleston & Kellermanns, 2004; Gersick et al., 1997).
Therefore, a greater number of family owners who have the formal right to voice their opinions will lead to weakened social relationships in family businesses. The lack of social cohesion is particularly harmful to owners who are also family members because it may lead to conflicts that spill over into other aspects of their lives (Kellermanns & Eddleston, 2004). A vicious circle can be established such that tensions permeate both family and business environments. Indeed, weaker ties and less social cohesion as a result of higher ownership dispersion lead to poor information flow, which may not only inhibit potential gains but also provoke counterproductive behaviors, such as path dependency and difficulties in challenging established strategic directions. Thus, when firms have many family owners and weak social ties, a situation can emerge that potentially paralyzes decision making and threatens firm.
It follows that greater ownership dispersion can decrease the probability of internal ownership transition through a number of different mechanisms related to embeddedness and structural cohesion. This includes the inability to agree upon who would be a suitable heir, conflicts between incumbents and potential successors, lack of trust and confidence in the potential successor, and lack of commitment of family members to the potential successor (De Massis et al., 2008). The sale of the business to an external party can be a way of responding to weaker ties and lack of social cohesion among family members. If nothing else, harvesting the value of the business and distributing it to family owners reduces the need for future interactions among parties. Taken together, the above discussion suggests that because of weaker relationships and social cohesion, greater ownership dispersion is associated with lower probability of within-family ownership transition. This leads to our first hypothesis:
Hypothesis 1: The probability of internal ownership transition decreases (compared to the probability of external ownership transition) with increased ownership dispersion.
Number and Age of Potential Heirs
Just as with the dispersion of ownership among the incumbent generation, the number of potential owners (i.e., heirs) in the next generation is likely to influence family firms' choice between internal or external ownership transitions. A larger set of potential heirs leads to a larger and more complex family structure. Because the family system overlaps with the business system (Harvey & Evans, 1994), such family complexities are likely to spill over into the business (McKee et al., forthcoming). For instance, compensation to those who do not take over ownership of the firm becomes more complex the larger the number of potential heirs is (Gersick et al., 1997), and the larger the number of potential heirs, the weaker the social ties between them (Arregle et al., 2007). Further, it is important to consider the potential heirs' age, particularly to what extent the potential heirs have reached adult age or not. This is an important consideration because younger heirs are less likely to have created independent career plans or opinions about the firm's development. They are more likely to maintain strong ties with their parents, support their parents' views, and have fewer independent or conflicting opinions (Hoy & Sharma, 2010).
The existence of young and adolescent potential heirs implies that the incumbent generation can foster the belief that it may be possible to create a legacy by keeping the firm within the family (provided this is their preference). There is still uncertainty regarding what members of the younger generation want to do as they become adults, but at least there is a possibility that they will be willing and able to engage in the family business in productive roles as they grow up. Therefore, the existence of young children in the family creates incentives for maintaining the business within the family at least up to the point when these children become old enough to make their own decisions as to whether they want to engage in the family business. Thus, the existence of young potential heirs creates incentives for a longer-term orientation concerning the family's involvement in the firm (Zellweger, 2007). As a result, if a transfer of ownership takes place in the presence of young potential heirs, it is more likely that the family will retain its ownership than if no young potential heirs are present.
This reasoning is supported by the socio-emotional wealth (SEW) perspective (Gomez-Mejia et al., 2007; Berrone, Cruz & Gomez-Mejia, 2012). If owner-families with young potential heirs have a stronger long-term orientation, it is more likely that they feel a stronger affective and emotional commitment to their business. Young heirs who have not yet embarked on independent professional careers tend to embrace their parents' viewpoints and priorities and thus also their concern for protecting the family's SEW endowment associated with its firm through an internal transition of ownership (Berrone et al., 2012).
As a way to increase the understanding of the role of SEW in the next generation, it is common in owner-families with many potential young heirs to introduce succession planning practices that increase the likelihood of internal ownership transitions. Examples of such practices are family councils, family policies, and family constitutions, which are used to support the family in communicating and building cohesion through agreement on shared values, goals, and visions (Lambrecht & Lievens, 2008; Hoy & Sharma, 2010). Thus, based on the above, we formulate the following hypothesis:
Hypothesis 2a: The larger the number of young children that can become potential heirs, the higher the probability of internal ownership transition (compared to the probability of external ownership transitions).
However, if the potential heirs have reached adult age, their roles in the family and the family business are likely to be different, and there is less uncertainty regarding their preferences and abilities. The existence of adult potential heirs who have not yet assumed ownership responsibilities increases the likelihood of tension within the family with regard to who should take over the firm. Adult family members are more likely to have formed career plans (inside or outside the family business) and have a wider set of network ties beyond the family. To the extent that they have been involved in the family firm, they are likely to hold independent opinions regarding firm development and to voice these opinions. The larger the number of adult potential heirs, the more difficult it will be to select the new owners from within the family. Assuming members of the older generation are aware of these possible cohesion problems resulting from a large group of potential new family owners, they may avoid these problems by instead transferring ownership externally.
When there are a larger number of adult potential heirs, family firms' embeddedness and their sharing of family/business values are likely to be less strong than in the case of young heirs, as is their concern for preserving the family's endowment of SEW (Berrone et al., 2012). In a situation with many potential adult heirs, it will be more challenging to implement the formal family governance practices that serve to unify family members, such as shared goals and visions, since different agendas may already exist within the family (Gersick et al., 1997). Further, the weaker embeddedness resulting from the existence of adult potential heirs who have not assumed ownership of the business may mean that the next-generation members have already chosen not to or have not been allowed to get involved in the firm. As a result, the family is not working closely together to advance the firm, and the expectations of keeping the firm in the family may not be as high as a result of the weakened cohesion in the family. Thus, adult potential heirs who are not involved in the ownership of the business may prefer different career paths than taking over the family firm. This leads to the following hypothesis:
Hypothesis 2b: The larger the number of potential adult heirs, the smaller the probability of an internal ownership transition (compared to the probability of external ownership transitions).
Intergenerational Ownership Involvement
The involvement of at least two generations in the business is a necessary requirement for a successful within-family transfer.. The involvement of the next generation implies that there is both an increased interest in learning the business and a willingness of potential successor(s) to take over the company (Venter et al., 2005; Le Breton-Miller et al., 2004). Hence, intergenerational ownership involvement in the family firm and older and younger generations working side by side can be part of preparing for ownership transfer to the next generation of family members.
In addition, the involvement of two generations can be a sign of social obligation–based family collaboration and lack of attractive opportunities outside the family firm (Sharma & Irving, 2005). These arguments are unlikely to hold for the Swedish context. In Sweden, individualism dominates. Family relationships are usually characterized by relatively weak family ties as opposed to family relationships found in Southern Europe, Latin America, or Asia, where the family has priority over the individual (Reher, 1998). Hence, when multiple generations work together, it can be a signal of the younger generation's affective rather than normative commitment to take over the company (Sharma & Irving, 2005). Affective commitment of successors contributes to creating strong social solidarity, which is reinforced by the creation of a joint understanding and shared agenda for how the business should develop and be run.
As ownership transition is a process rather than an event, the older founder generation collaborating with the younger generation is a crucial part of an efficient transition within the family (Sorenson, 1999). For instance, the coexistence of multiple generations enables the older generation to build stronger ties and involve the younger generation in creating the long-term goals of the family business as well as to transfer their tacit business knowledge to the younger generation (Miller & Le Breton-Miller, 2006). The involvement of multiple generations is naturally in line with the family's priorities, such as transmitting the family's business culture and identity (Dyer, 2006), including a long-term orientation, to the younger generation (Miller & Le Breton-Miller, 2006). Hence, the intergenerational transmission of job-related norms and values to the younger generation creates stability within the owner-family (Arregle et al., 2007). This stability is necessary for strong social relationships to emerge and therefore creates strong cohesion among the owner-family.
Further, working together enables both generations to generate a shared vision and agenda of the firm. .This shared vision ensures a certain degree of connectedness and solidary among the family members (Bjuggren & Sund, 2001a). Such a connectedness is crucial for the older generation to feel confident enough to hand over the business to the younger generation and also for the different individual members within the younger generation to feel they are in agreement with each other. Thus, collaborating allows both generations to develop a better understanding of each other and provides them the opportunity to adjust to and respect each other's views and expectations (Sorenson, 1999). Multiple generations working together increase the family's cohesion. This mechanism of adjustment enables family members to strengthen their relationships, which results in a stronger embeddedness of the firm within the family (Handler, 1990). Thus, we expect that when multiple family generations work together and collaborate in their firms, social relations will strengthen, encouraging a strong feeling of togetherness. This strong emerging cohesion will lead to increased willingness to keep the firm within the family, which leads us to our third hypothesis:
Hypothesis 3: The probability of internal ownership transition increases (compared to the probability of external ownership transition) when multiple generations are involved in the business.
Family or Non-Family CEO
From our theoretical perspective, we expect that whether the family firm is operated by a CEO from the owner-family or an external CEO influences the choice between internal and external ownership transition. Appointing an internal family CEO can be regarded as a sign of strong interest and commitment to develop the firm among the family members involved in the business (Hall & Nordqvist, 2008) and tends to be associated with a wish to continue to operate the family business in line with the family's values, experiences, and traditions (Bjuggren & Sund, 2001b). Although it has been posited that families may choose to retain long-term ownership despite appointing an external CEO (Bergfeld & Weber; Salvato et al., 2012), these results are based on the investigation of large to very large and mainly publicly listed companies, while our focus here is primarily on small- to medium-sized firms.
From the social cohesion perspective, having a family member as CEO shows that the family has selected a leader who is embedded in the family and who represents continuity in terms of the family's agenda with the firm. In other words, assigning a family member as the main decision maker signals that there is mutual consent within the family to limit external influences in the company (Arregle et al., 2007). Appointing a family CEO is also a symbol of the owner-family's stability and durability (Stewart & Hitt, 2012) because the CEO is a member of the family group and is assumed to better protect the owner-family's interests. Further, the appointment of a family CEO increases the interdependence of family members (Arregle et al., 2007). Such a CEO is more likely to pursue the family's goals and interests and work towards creating a cohesive owner-family that supports a shared agenda. This notion is grounded in the view that an internal ownership transition can be regarded as a process, through which executive leadership is transferred from one generation to another as a first step towards transitioning ownership within the family. Indeed, retaining control within the family has been identified as a fundamental goal of many family businesses (Lee, Lim, & Lim, 2003; Schulze et al., 2003), and appointing the CEO is an important part of securing this control.
Hiring an external CEO may imply the absence of a suitable internal leader or a disagreement in the family regarding what strategic direction the firm should take (Chua, Chrisman & Sharma, 2003). There are several reasons why a family may not agree on an internal family member as the CEO and why appointing an external CEO is often a sign of a forthcoming sale of the firm to outside owners (Bjuggren & Sund, 2001b). First, the family may not be able to reach consensus regarding which family member should be appointed as CEO. Appointing an external CEO becomes the only way to deal with this issue and indicates potentially weak stability and solidarity within the family. Second, there may be little interest and willingness among the next generation of family members to devote the time and energy needed to take over the business (Dyer, 1989), including the stressful work of dealing with expectations and the different opinions of family stakeholders. Alternatively, the social relations between family members may already be characterized by weak cohesion and disagreements, and appointing an external CEO could be seen as the only viable option to keep the firm well managed (Gersick et al, 1997). Third, the family may have no intention to continue running the firm in the future and stay as owners to perpetuate the family business's values and visions (Dyer, 1989). Here, appointing an external CEO tends to be a first sign of the diminishing influence of the family on the business, eventually leading to an external ownership transition.
While appointing an external CEO may be a deliberate step by the family towards a sale of its business, once appointed, the external CEO may also have little interest in preserving the business as a "family business" since he/she is emotionally less attached to the business and is more concerned with the firm's financial performance. When an external CEO does not have an interest in pursuing the family's goals, which are both financial and non-financial (Hall & Nordqvist, 2008), this could lead to an even weaker embeddedness and structural cohesion among the family members, decreasing the level of intra- group solidarity within the family. As an increasingly unstable owner-family is associated with negative emotions that can have a destructive impact on the family business (Jehn, 1997), leading to an external ownership transition. Taken together, these arguments lead us to the following hypothesis:
Hypothesis 4: The probability of internal ownership transition decreases (compared to the probability of external ownership transition) with the appointment of an external CEO.
METHODS
Research Design and Sample
Examining and contrasting firms transferred within families and firms transferred to external owners posed several methodological challenges. First, we needed to identify a relevant sample of family firms. Second, we needed to determine when a transfer took place and under which conditions the transfer could be considered an internal or external transfer of ownership. Third, we needed to consider relationships between individuals as well as firm-level aspects.
In order to deal with these potential challenges, we constructed a longitudinal dataset by combining five longitudinal databases maintained by Statistics Sweden, the official census bureau in Sweden. The RAMS and CFAR databases provide yearly data on all firms registered in Sweden, including measures like sales turnover, profitability, and debt. The databases LISA and Jobbregistret (work register) provide yearly data on all Swedish inhabitants, family relationships, and income sources. Finally, the multigenerational database provides information on couples (i.e., if they are married, living together, and have children together) as well as on biologically linked families (i.e., parents, children, and siblings).
All of these databases contain annual information about individuals and/or firms. Thus, our sample and analyses are based on annual data observations. As a sampling frame, we chose all privately held firms with 10 employees or more that were in existence in Sweden in 2004 and followed them until 2008. Thus, we excluded smaller family firms that may not be realistic acquisition targets. Among these firms, we included all firms that were owned by two or more family members either in a household (i.e., spousal couple) or in a biologically linked family (i.e., fathers, mothers, children, and siblings). We base our study on information obtained from a five-year period (2004 to 2008). However, since we use information in 2004 to predict ownership changes taking place between 2004 and 2005, we end up with a four-year panel with a lagged dependent variable.
We followed all family firms and recorded ownership changes for each consecutive year. From one year to another, a firm could either (1) remain intact without any ownership transition, which we labeled continuation; (2) go through an internal ownership transition; (3) go through an external ownership transition; or (4) experience a firm dissolution, which is labeled shutdown. The result is a sample of 3,829 family firms, corresponding to 12,125 firm-year observations. The average firm had 25.59 (s.d. 40.26) employees and earnings before interest and tax (EBIT) of 2,184,125 SEK (approximately 322,000 USD) (s.d. 7,411,496 SEK).
Variables and Measures
Dependent variable. Ownership transition. Our dependent variable can take two values reflecting ownership transitions: internal transition or external transition (Wennberg et al., 2011). We observe ownership transitions based on comparing the ownership states of adjacent years. If at least one owner of a family firm completely exited ownership from one year to the next, this is considered an ownership transition. If a spouse, child (or several children), or sibling(s) either remained as owners or entered ownership in the same year, this was considered an internal ownership transition, and the variable was coded 1. If all owners of a family firm exited ownership from one year to the next and new owners who do not belong to the same core family as the prior owners entered during the same time period, this was considered an external ownership transition and was then coded 0. In order to ensure that our results were not biased, we accounted for all other possible outcomes in our estimations. First, if a firm went through no ownership change during a given year, the family firm is retained in the sample and the variable was coded 2. Second, if the firm experienced a shutdown at any time during the studied time period, the firm was retained in the sample until the year it shut down, and the variable was coded 3 in that year. With these definitions, during the four potential ownership transition periods covered by our panel, approximately 56.4% of all firms belonged to the continuation category, 23.5% of all firms experienced a transfer within the family, 17.4% of the firms had transitions to outsiders, and 2.7% of all firms belonged in the shutdown category.
Independent variables. Ownership dispersion was measured as the total number of owners (similar as Schulze et al., 2003). By Swedish law, individuals must report their activities in a closely held company to the tax authorities. Consequently, all individuals who filed taxes for their ownership in this kind of company form were considered part owners. Ownership dispersion is an important indicator for the social cohesion within the owner family (cf. Moody and White, 2003). A higher number of potential young heirs was measured as the total number of children aged 17 years or younger of all owners. These children are likely to still live at home, and they are not yet considered adults in the Swedish legal system. Because the variable was heavily skewed, it was log transformed (with 1 added to all cases to avoid missing values). The number of potential adult heirs was measured as the total number of children 18 years of age or older of all owners. The total number was corrected for spousal couples who are both part owners and have mutual children. This variable was also log transformed. Multiple generations indicates whether a parent owns the firm together with his/her child (or children) or not (cf. Le Breton-Miller et al, 2004). When only one generation owns the firm, the variable was coded 0; otherwise, it was coded 1. Internal CEO measures whether the CEO is also a family member (coded 1) or not (coded 0) (cf. Salvato et al., 2012; Anderson & Reeb, 2003).
Control variables. We controlled for several potential influences on ownership transitions. We captured past performance in terms of EBIT (earnings before interest and tax as reported to the tax authorities), which was entered in log format. Poor performance could create incentives for selling the business to an external party, especially if the business is in financial distress (Wennberg et al., 2010). Another indication of performance is owner mean salary, measured as the average salary of the owners. This is an indication of the ability of the family business to provide an attractive economic situation for its owners. The greater the mean salary of the owners, the more likely that ownership transfer will be internal. This variable was log transformed.
To capture future orientation, we measured financial leverage. Family firms that are managed for the long haul with the intention to remain within the family's control for generations are likely to possess fewer leveraged capital structures that minimize the risk of losing control rights to ensure the firm stays out of financial distress. Limiting the risk of losing control rights can be achieved by enforcing a pecking order among sources of capital (Myers, 1984), avoiding external debt, and minimizing leverage (Mishra & McConaughy, 1999). Thus, we could expect family firms that are to be retained under family control to be more likely to forego potentially profitable investments if they were to be funded by increasing debt, whereas we expect family firms intended for external ownership transfer to be more willing to assume greater debt. Financial leverage was measured as the percentage of total liabilities in proportion to owners' equity. Owner mean age was measured as the average age of the owners. To some extent, this variable taps ownership lifecycles, and a high owner mean age is likely to increase the likelihood of ownership transitions (Harveston et al., 1997).
Owners do not have to work in the firm or even have it as their main source of income. Therefore, we measured full-time owners in percent by dividing the number of owners who obtain their main income from the firm by the total number of owners. A lower percentage indicates that there are owners who have some other occupation as their main source of income. Another variable related to family structure and the strengths of ties within the owner-family is non-mutual heirs. This variable measures the number of children with one partner of the spousal couple as a parent but not the other. A larger number of non-mutual heirs would indicate weaker relationship ties and therefore a higher probability of an external ownership transfer. In order to control for if management succession coincides with ownership transitions, we included the control variable CEO exit, which indicates if the CEO changes from one year to another (coded 1) or not (coded 0).
We also relied on a number of traditional control variables. Firm size was measured as the total number of employees. Since the variable was skewed, we used the logged value in the analyses. We controlled for industry groups by using a set of 15 dummy variables (similar to Smith & Amoako-Adu, 1999). Year dummies were constructed for the years 2005, 2006, and 2007. The year 2004 served as the comparison year. Firm Age is a dummy variable coded 0 for firms 15 years or younger and coded 1 for older firms.
ANALYSES AND RESULTS
Descriptive statistics and correlations are presented in Table 1 (year and industry controls suppressed). Given that we study the full population of family firms, the descriptive statistics are worthy of comment. The mean of ownership dispersion (H1) is 2.8, indicating that most family businesses with 10 or more employees have concentrated ownership, with less than three owners on average. Examining the variable multiple generations (H3), the number 0.34 indicates that only one-third of the firms are owned simultaneously by multiple generations. Combined, these numbers suggest that spousal couples and/or siblings are the more common ownership constellations. The mean numbers of potential young heirs (H2a) and potential adult heirs (H2b) are 2.0 and 2.5, respectively, suggesting that it is more common for family business owners to have adult children who are not (yet) heirs than young children. To a large extent, this can be explained by the mean age of the ownership team, which is close to 50 years. Not surprisingly, as can be seen in Table 2, there is a relatively strong negative correlation between owner mean age and potential young heirs (r = -0.471 , p < 0.001). The variable internal CEO has a mean of 0.73. This suggests that close to three-fourths of all family firms rely on a CEO who is also a family member.
In terms of the control variables, we note that the variable full-time owners has a mean of 0.92. This reflects that 92% of all owners also have the family firm as their main source of income (and an even higher share of all family firms have at least one family member who works in the firm). This suggests that owners who are only passive investors are uncommon.
========================================
INSERT TABLES 1 & 2 ABOUT HERE
========================================
In order to avoid potential bias we rely on a multinomial logit model taking into account firms that do not experience any ownership transition or are shut down during the studied period. If a firm went through no ownership change during a given year, the family firm is retained in the sample. If the firm experienced a shutdown at any time during the studied time period, the firm was retained in the sample until the year it shut down. Other potential ownership transitions, such as mergers and acquisitions, were excluded from the sample and from the analyses. With these definitions, during the four potential ownership transition periods covered by our panel, approximately 56.4% of all firms belonged to the continuation category, 23.5% of all firms experienced a transfer within the family, 17.4% of the firms had transitions to outsiders, and 2.7% of all firms belonged in the shutdown category. Table 3 tests the hypotheses, comparing internal and external transfers, while suppressing the categories continued and shutdown firms to enhance readability. Following prior research (e.g., Wennberg et al, 2010) we constructed a pooled sample with yearly indicator using the cluster option in STATA and robust standard errors .To deal with potential multicollinearity we estimated value inflation factors (VIFs), which were all below 3, suggesting it was not a problem. As a further robustness test, we also entered the highly correlated variables independently. These estimations generated substantively identical results. Thus, multicollinearity is not a threat to the validity of our findings.
Our base model containing the control variables is displayed as Model 1 in Table 3. Statistically significant effects are noted for mean age of the owners (positive), year dummy 2005 (negative, suppressed from the table), mean age of the owners (positive). Model 2 in Table 3 adds the variables corresponding to the hypotheses. There is a positive and statistically significant effect of ownership dispersion on the probability of internal ownership transition (0.154, p < 0.01). This effect is opposite to what we anticipated in Hypothesis 1, which stated that the higher the ownership dispersion of a family business, the lower the probability of within-family ownership transition.
According to Hypothesis 2a, we expect that the number of potential young heirs (17 years or younger) will increase the probability of an internal transfer. The coefficient for young heirs has a positive but statistically non-significant effect on the probability of internal transfers (0.06, p > 0.05). Thus, Hypothesis 2a is not supported. There is however a negative and statistically significant effect for the number of adult potential heirs on the probability of intra-family ownership transition (-0.259, p < 0.05). This result supports Hypothesis 2b, which stated that the larger the number of potential adult heirs. the smaller the probability of an internal ownership transition. The coefficient for multiple generations has a positive and statistically significant impact on internal transfers (1.919, p < 0.001). These results lend support for Hypothesis 3, which anticipates that the probability of internal transfers (compared to the probability of external transfers) will increase when multiple generations are involved in the business.
Finally, Hypothesis 4 posits that the probability of internal ownership transitions (compared to the probability of external ownership transitions) will decrease with the appointment of an external CEO. The coefficient for internal CEO is positive and statistically significant (1.349, p < 0.001), meaning that having an internal CEO increases the probability of internal transfers. Put differently, the existence of a non-family CEO decreases the probability of internal transfers. Thus, our results support Hypothesis 4. In sum, our results support Hypotheses, 2b, 3 and 4.
Because we were surprised to find that the testing of Hypothesis 1 led to the reverse result of what we hypothesized, we decided to explore whether a curvilinear relationship exists between ownership dispersion and the probability of within-family ownership transitions, so we conducted a post hoc test. To do so, we added a square term of the ownership dispersion variable to the equation. The results retained the positive and significant effect of ownership dispersion, but the coefficient of the squared term was negative and significant (-0.004; p < 0.01). This suggests an inverse U-shaped relationship between ownership dispersion and the probability of internal ownership transition, which implies that a low to moderate number of owners is positively associated (0.269, p < 0.001) with internal transfers, while a high level of ownership dispersion (corresponding to a larger influence of the squared term) has a negative effect (-0.005, p < 0.01) on internal transfers. Thus, we find partial support for Hypothesis 1.
DISCUSSION
Rather than assuming that internal ownership transition is the preferred choice and external ownership transfer a last resort, we have taken the view that ownership transition in family firms can be seen as a choice. Some family firms choose to transfer ownership inside the family to the next generation of family owners, whereas other family firms choose to sell their firms to new owners outside the family. We found that 23.5% of the firms studied were transferred within the family, while 17.4% were transferred to outsiders (during the same time period, 2.7% were shut down and 56.4% continued operations without ownership changes). Approximately 40 % all the family businesses studied went through ownership changes during our four-year window of observation. Although internal transfers are somewhat more common than external transfers, a large share of ownership transfers were external during the time period we studied. Overlooking these external transfers or viewing them as simply a failure to retain the business in the family is likely to lead to biased inferences in our understanding of ownership transitions and strategic choices in family businesses (Zellweger et al., 2012).
Drawing on the embeddedness perspective and the notion of structural cohesion, we hypothesized and tested how several factors related to the structure of the family were related to the probability of internal and external ownership transfers. We hypothesized (Hypothesis 1) that greater ownership dispersion would be associated with a lower probability of internal ownership transfer. Interestingly, we found an inverse U-shaped relationship between ownership dispersion and probability of internal ownership transfers. Initially, greater ownership dispersion actually increased the probability of internal transfer, but beyond a certain level, greater ownership dispersion increased the probability of external transfer. This finding provides some support for our idea that as the group of family owners becomes larger, it becomes more difficult to maintain strong ties and solidarity and thus to agree on a shared agenda for business development and governance. In this situation, conflicts may emerge, and the alternative of selling to outsiders becomes relatively more attractive (Eddleston & Kellermanns, 2007). However, this effect only seems to become present beyond a certain level of ownership dispersion. At low levels, increased ownership dispersion actually increased the probability of internal ownership transfer. One explanation could be that families committed to internal ownership transition pass on some ownership to members of the next generation early on to get them to commit to the business. We conducted some post hoc analyses to test if we could find any evidence of such behavior within the four-year window of our study, but none was found. As we know from prior research (Le Breton-Miller et al., 2004; Sharma et al., 2003), succession is a process that takes time. Apparently, the four-year time frame of our study was insufficient to capture this process in our sample.
Another reason for this finding may be related to the definition of our categories as internal and external ownership transitions. In order to discriminate family from non-family firms, a demarcation must be inserted somewhere (Dyer, 2006; Sharma, 2004). This demarcation will always be somewhat arbitrary because firms represent a continuum ranging from purely family to purely non-family businesses, while others fall somewhere in between. The same considerations occur in our research when determining if family businesses remain family businesses after an ownership transition. For example, we do not capture ownership transitions to remote relatives in the internal category. Such choices can potentially influence the results.
An interesting implication of ownership dispersion and transition of ownership relates to the long-term prospects of family businesses. Each time a family business is transferred from one generation to the next, family ownership is likely to become increasingly dispersed (Sharma et al., 2003). Given our findings, it should also become increasingly more likely that ownership is transferred to external owners as a result of the weaker ties and cohesion within the family. This could be an important explanation as to why so few family businesses remain within a family for multiple generations. Even in our sample consisting of firms with concentrated ownership (mean of 2.8 owners per firm), external transfers were almost as common as internal transfers.
Next, we hypothesized that the number of young (below 18 years of age) and adult (18 years or older) potential heirs would have opposite effects on internal and external ownership transitions. Only the hypothesis 2b was supported. It thus appears that number of adult heirs better predicts the choice between internal and external ownership transitions than young heirs. Nevertheless, the separation of young versus old potential heirs has, to our knowledge, not been examined before. Our research heeds the call for taking into account the family members' life stage to understand the choice of ownership transition type (Hoy & Sharma, 2010). We measure number of young heirs as the total number of children that are 17 years or younger. It is plausible that the impact of potential young heirs on the choice between internal and external ownership transition differs depending on whether the heirs are small children, young teenagers or approaching adolescence.
The finding that a higher number of adult potential heirs decreases the likelihood for internal ownership transfer, although the pool of potentially competent heirs is larger, while young heirs has a positive non-significant effect, may indicate that the life stage of potential heirs is an important dimension when understanding ownership transitions in family firms. It is plausible that the potential future involvement of the next generation as owners creates different dynamics with regards to embeddedness and cohesion compared to the dynamics between de facto owners. With the existence of young potential heirs, there is a possibility that they will take over the firm as they grow up. We anticipated that this would create incentives for maintaining the business within the family at least until these young heirs grow up, facilitating a longer-term orientation concerning the family's involvement in the firm (Zellweger, 2007). A fundamental aspect of parenting is making long-term plans for children when they are young. However, these dynamics may change as children grow up and begin forming their own opinions, values, and independent career plans. The fact that we have grouped very young children with adolescent children could imply that we miss some of these dynamics.
Our results regarding the negative effects of adult heirs supports the idea that a larger pool of potential adult heirs leads to weaker ties, less solidarity, and less concern for preserving the family's endowment of SEW among the potential heirs but also leads to a larger pool of candidates who may be willing and able to take over ownership of the firm. Our findings suggest that the negative aspects of potentially weaker embeddedness and cohesion outweigh the potentially positive aspects of greater competence and motivation. This is important for future theorizing about the drivers and consequences of ownership transitions in family firms. In addition, a frequent argument put forth in family business research is that the pool of potential heirs may be too small and thus prevent intra-family ownership transition. Interestingly, our findings concerning the negative influence of a larger pool of adult potential heirs seems to be at odds with this argument.
Taken together, our findings highlight the value of family governance as a way of building cohesive owner-families. An arsenal of methods and mechanisms for dealing with ownership transition in situations like these exists, including family councils, family protocols, specific policies, and other formal governance structures (Gersick et al., 1997; Schulze et al., 2003). Despite these mechanisms, family businesses may not be aware of or use them, or they may not be sufficiently efficient in facilitating a smooth ownership transition to the next generation. If they were, we may not have found the negative implications of the number of potential adult heirs that we observed.
Hypothesis 3 was also supported. We found that internal ownership transition within the family is more probable than external ownership transition when multiple generations are involved in the business. This finding is in line with our theoretical argument that when two or more generations of the family jointly own the firm, they are more likely to develop an intergenerational perspective of their involvement in the firm and create a vision that embraces the choice of keeping the firm within the family. In this situation, as part of creating a shared perspective and vision, both generations are able to intentionally work on building strong ties and solidarity between family members (Sorenson, 1999), thus increasing the likelihood of an internal transition of ownership.
Testing the first three hypotheses conjointly some normative implications emerge. For a family that plans to retain ownership within the family and to pass it on to coming generations, it seems that it is a good idea to pass at least some ownership on to the next generation while they are still young. Ownership across multiple generations was positive for internal transfer, whereas adult children who did not have an ownership stake in the business decreased the probability of internal transfer. Further, at reasonably low levels, ownership dispersion also positively influenced internal transfer. Our results give empirical support to previous suggestions in the literature that planning for succession should start early and that it is a process that takes time (LeBreton-Miller et al., 2004; Sharma, 2004).
Regarding Hypothesis 4, we found that firms with a CEO from the owner-family are more likely to be passed on to other family members than firms with an external CEO. This supports our general theoretical assumption that with strong family presence in the highest executive decision-making position of the firm, firms are characterized by more commitment to family values and interests in strategic choices than firms with an external CEO. Having a family CEO is also a way to select one family member who has the authority to navigate business development even if ties and structural cohesion is weakening between owner-family members. While an external CEO from outside the family can serve as a non-family mediator among family stakeholders (Blumentritt, Keyt & Astrachan, 2007), appointing an external CEO often implies less intensive family commitment to firm operations. In many cases, this is the first step to selling a firm to new non-family owners (Bjuggren & Sund, 2001b), while in other cases—mainly in larger family businesses—appointing a non-family CEO can be the result of the owner-family moving from an owner-operator mode to an owner-investor mode (Habbershon & Pistrui, 2002). Our findings give further empirical support to this observation. We even suggest that in many firms, these two dimensions can coincide. In other words, family members appoint an external CEO both as a way to deal with or as a result of an increasingly unstable owner-family and as way to prepare the business for sale to outside owners.
Limitations and Future Research
Our article contains limitations, all of which represents important avenues for future research. One limitation is that we do not include merged and acquired firms in our study, so we do not distinguish between different kinds of acquirers when examining external ownership transfers. For example, it would be interesting to know if the acquirer is a private individual, a competing firm, or a private equity firm. It could be that the variables influencing family firms' preference for internal or external transition of ownership differentially influence the different types of external transitions. Further, our model takes firm closures into account. Thus, bankruptcies and other forms of involuntary closures do not bias our results. However, more fine-grained types of failures (e.g., the incapacitation of the current leader) were not considered.
Future research should also focus on the shift in mindsets among many business families from seeing themselves less as owner-operators and more as owner-investors (Habbershon & Pistrui, 2002). This shift in mindset may imply that owner-families increasingly remove themselves as managers in their firms but stay as active owners. A limitation of our research is that we were not able to take this shift into account specifically.
A further limitation is that we do not have information on different family members' exact ownership shares. As a result, we do not capture ownership transitions that involve a re-distribution of shares owned by a family. Another potential limitation relates to our definitions of family firms and of internal ownership transitions. We limited family firms to those firms that included ownership by parents, children, siblings, and couples and limited internal family transition to only include ownership transfers to spouses and children. Thus, we excluded firms owned by cousins, for example, if no other closer family members were included and also excluded ownership transfers to siblings or cousins.
A non-finding also triggers ideas for future research. We examined performance (EBIT and owner salaries) as well as leverage, but we found no direct effect for either. Although we believe these variables are important for understanding ownership transitions, relationships may be more complex. Many family business owners have different reference points because of concerns for the SEW they derive from maintaining family control over a firm (Berrone et al., 2012). Therefore, a given level of firm earnings and private remunerations likely influence owner-families differently. Further, families have access to substantial information about their businesses that is not easily accessible for outsiders. This information asymmetry can be used opportunistically, which creates incentives for "window dressing" firms presented as potential acquisition targets to outsiders—namely, firms exhibiting artificially high performance prior to external ownership transfer (Wennberg et al., 2012). Expectations of future performance are also likely to influence whether a business will be passed on to the next generation. A more fine-grained examination of how performance influences the decision between internal and external ownership transfer would be valuable.
A final limitation is that our key argument about the influence of embeddedness and cohesion of owner-families on their choice of voluntary ownership transition is inferential. We used this theoretical perspective to identify relevant variables, but we did not test the strength of ties and cohesion nor solidarity between family members directly, nor were we able to directly test if the ownership transitions we observed were all voluntary. That is, because of our use of secondary data, we did not have access to direct measures in the different businesses. The fact that we used this framework to identify the selected variables and develop our hypotheses and that our hypotheses were mostly supported speaks in favor of our interpretation of the results. However, we cannot rule out that alternative causal mechanisms are at play and that other theoretical perspectives may be useful to understand family firms choice between an internal or external ownership transition.
Implications for Research
Our research has implications for the entrepreneurship and family business literatures. In particular, we extend our understanding of the drivers and consequences of ownership transitions in family firms. Most previous research has focused on management succession rather than on ownership transition. While these two separate processes sometimes go hand in hand, ownership transition is in itself common and complex enough to require more research. In particular, this article suggests that research benefits from not assuming internal ownership transition is always preferred for families operating businesses but rather views different modes of ownership transition (i.e., internal, external, and possibly others) as voluntary choices. This is important because the failure to recognize outside ownership transition options and excluding family firms that opt for such an ownership transition in empirical studies may lead to biased results in a fashion similar to the bias associated with sample selection. In particular, our study shows how owner-family structure and involvement impact whether family firms in an entire population—not just a sample—undergo internal or external ownership transition.
Our research also has implications for the family perspective of family firms. A central assumption in the family business field is that the business is embedded in the family and the family in the business. Both scholars and practitioners focusing on family firm dynamics and transitions have suggested that to better understand key family firm choices, we should address the intertwined nature of the family and the business (James et al., 2012). We used an embeddedness perspective and the notion of structural cohesion (Granovetter, 1985; Moody & White, 2003). This allowed us to develop a novel conceptual framework to predict how important and tangible factors of family structure and involvement (i.e., family ownership dispersion, number and age of potential heirs, intergenerational ownership involvement, and whether the CEO is a family or a non-family member) impact the type of ownership transition chosen.
For instance, our findings suggest that greater ownership dispersion and a larger pool of potential adult heirs are associated with a higher probability of external ownership transition relative to internal ownership transition. Thus, the more present owners and/or potential future owners are involved, the greater the probability of external ownership transfer. Interpreted within the conceptual framework developed here, we suggest that it does not lead to a "failed ownership transition"; rather, it makes other choices more attractive. The conceptual framework based on embeddedness and structural cohesion seems to be robust and useful enough to be used to investigate and explain other important choices in family business contexts besides ownership transition. Therefore, our study extends our general knowledge about how embeddedness and family cohesion through owner-family structure and family involvement in the firm have implications for family firms' key choices (Sharma & Manikutty, 2005; James et al., 2012).
We also believe that our approach with the variables (e.g., the number of potential heirs, family ownership structure, intergenerational ownership involvement, and the appointment of an internal family CEO) representing indicators of an owner-family's structural cohesion and thus their inclination to choose either an internal or external ownership transition is promising and deserving of following. Such variables are relatively easy to assess compared to perceptive measures of cohesion and/or embeddedness.


REFERENCES

Aldrich, H. E. & Cliff, J. E. (2003). The pervasive effects of family on entrepreneurship: Toward a family embeddedness perspective. Journal of Business Venturing, 18 (5), 573-596.
Anderson, R. C. & Reeb, D. M. (2003). Founding-Family Ownership and Firm Performance: Evidence from the S&P 500, The Journal of Finance ,58 (3), 1301-1328.
Arregle, J.-L., Hitt, M. S., Sirmon D. G., & Very, P. (2007). The development of organizational social capital: attributes of family firms. Journal of Management Studies, 44 (1), 73-95.
Bergfeld, M.-M. & Weber, F.-M. (2011). Dynasties of innovation: highly performing German family firms and the owners' role for innovation. International Journal
Berrone, P., Cruz, C. C., & Gómez-Mejía, L. R. (2012). Socioemotional wealth in family firms: A review and agenda for future research. Family Business Review, 25( 3), 258-279
Bjuggren, P.-O., Sund, L.-G. (2001a). Strategic decision making in intergenerational successions of small- and medium-size family-owned businesses. Family Business Review, 14(1), 11-24.
Bjuggren, P. & Sund, L. (2001b). A transaction cost rationale for transition of the firm within the family. Small Business Economics, 19, 123-133.
Blumentritt, T.P., Keyt, A.D. & Astrachan, J.H. (2007). Creating an environment for successful nonfamily CEOs: An exploratory study of good principals. Family Business Review, 20(4), 321-335.
Carney, M. (2005). Corporate governance and competitive advantage in family-controlled firms. Entrepreneurship Theory and Practice, 29, 249-265.
Cohen, J. & Cohen, P. (1983). Applied multiple regression/correlation analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum.
Chua, J., Chrisman, J., and Sharma, P. (2003). Succession and nonsuccession concerns of family firms and agency relationship with nonfamily managers. Family Business Review, 16(2), 89–107.
De Massis, A., Chua, J., & Chrisman, J. (2008). Factors preventing intra-family succession. Family Business Review, 21(2), 183-199.
Dyer, W.G. (2006). Examining the 'family effect' on firm performance. Family Business Review, 19(4), 253-273.
Dyer, W.G. & Whetten, D.A. (2006). Family firms and social responsibility: Preliminary evidence from the S&P 500. Entrepreneurship Theory and Practice, 30(6), 785-802.
Dyer, W.G. (1989). Integrating Professional Management into a Family Owned Business. Family Business Review, 2(3), 221-235.
Eddleston, K.A. & Kellermanns, F.W. (2007), Destructive and productive family relationships: A stewardship theory perspective, Journal of Business Venturing, 22, 545-565.
Gersick, K.E., Davis, J.A., Hampton, M.M., & Lansberg, I. (1997). Generation to generation: Life cycles of the family business. Boston, MA: Harvard Business School Press.
Gomez-Mejia, L. R., Haynes, K. T., Núñez-Nickel, M., Jacobson, K. J. L. & Moyano-Fuentes, J. (2007). Socioemotional Wealth and Business Risks in Family-Controlled Firms: Evidence From Spanish Olive Oil Mills. Administrative Science Quarterly, 52(1), 106–137.
Granovetter, M. (1985). Economic action and social structure: The problem of embeddedness. The American Journal of Sociology 91 (3), 481-510.
Habbershon, T.G, Pistrui, J. (2002). Enterprising families domain: Family-influenced ownership groups in pursuit of transgenerational wealth. Family Business Review, 15, 223-238.
Hall, A. & Nordqvist, M. (2008). Professional management in family businesses: Extending the current understanding. Family Business Review, 11(1), 51-69.
Handler, W. (1990). Succession in family firms: A mutual role adjustment between entrepreneur and next-generation family members. Entrepreneurship Theory and Practice, 15(1), 37-51.
Harveston, P. D., Davis, P. S., & Lyden, J. A. (1997). Succession planning in family business: the impact of owner gender. Family Business Review, 10 (4), 373-396.
Harvey, M. & Evans, R.E. (1994). Family business and multiple levels of conflict. Family Business Review, 7(4), 331–348.
Hoy, F., & Sharma, P. (2010). Entrepreneurial family firms. Upper Saddle River, NJ: Prentice Hall.
James, A. E., Jennings, J. E., & Breitkreuz, R. S. (2012). Worlds apart? rebridging the distance between family science and family business research. Family Business Review, 20(10), 1-20.
Jehn, K.A. (1997). A quantitative analysis of conflict types and dimensions in organizational groups. Administrative Science Quarterly, 42(3), 530–558.
Jensen, M. C. & Meckling, W. H. (1976). Theory of the firm: managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3, 305-360.
Kellermanns, F. W. & Eddleston, K. (2004). Feuding families: When conflict does a family firm good, Entrepreneurship, Theory and Practice, 28(3), 209-228.
Kets de Vries, M.F.R. (1993). The dynamics of family controlled firms: The good and the bad news. Organizational Dynamics, 21, 59–71.
Lambrecht, J. & Lievens, J. (2008), Pruning the family tree: An unexplored path to family business continuity and family harmony Family Business Review, 21(4), 295-313
Le Breton-Miller, I, Miller, D, Steier, L., (2004). Toward an integrative model of effective fob succession. Entrepreneurship Theory and Practice, 28(4), 305-328.
Lee, K.S., Lim, G.H., & Lim, W.S. (2003). Family business succession: Appropriation risk and choice of successor. Academy of Management Review, 28(4), 657-666.
Long, J. S & Freese, J. (2006). Regression models for categorical dependent variables using Stata, Second Edition. College Station,TX: Stata Press.
Long, R. & Chrisman, J.J. Management Succession, In: Melin, L., Nordqvist, M. & Sharma, P. Sage handbook of family business, London: Sage Publications, forthcoming
McKee, D'L, Madden, T.M. Kellermanns, F.W., & Eddleston, K. A. Conflict in family firms: The good and the bad, In: Melin, L., Nordqvist, M., & Sharma, P., Sage handbook of family business, London: Sage Publications, forthcoming
Melin, L. & Nordqvist, M. (2007), The Reflexive Dynamics of Institutionalization: The Case of the Family Business, Strategic Organization, 5 (4), 321-333
Miller, D. & Le Breton-Miller (2006). Family Governance and Firm Performance: Agency, Stewardship and Capabilities. Family Business Review 19(1), 73-87.
Miller, D., Steier, L., & Le Breton-Miller, I. (2003). Lost in time: intergenerational succession, change, and failure in family business. Journal of Business Venturing, 18(4), 513-531.
Moody, J. & White, D. R. (2003). Structural cohesion and embeddedness: A hierarchical concept of social groups. American Sociological Review 68(1), 103-127.
Reher, D. S. (1998).Family Ties in Western Europe: Persistent Contrasts. Population and Development Review 24(2), 203-234.
Salvato, C., Minichilli, A., & Piccarreta, R. (2012). Faster Route to the CEO Suite: Nepotism or Managerial Proficiency? Family Business Review 25 (2), 206-222.
Sardeshmukh, S. R. & Corbett, A. C. (2011). The duality of internal and external development of successors. Opportunity Recognition in Family Firms, Family Business Review, 20 (10): 1-15.
Schulze, W.S., Lubatkin, M. H., Dino, R. N. & Buchholtz, A.K. (2001), Agency relationships in family firms: theory and evidence, Organization Science, 12, 99-116
Schulze, W. S., Lubatkin, M. H., & Dino, R. N. (2003). Exploring the agency consequences of ownership dispersion among the directors of private family firms. Academy of Management Journal, 46, 179-194.
Sharma, P., Chrisman, J.J., & Chua, J.H. (2003). Predictors of Satisfaction with the Succession Process in Family Firms. Journal of Business Venturing, 18(5): 667-687.
Sharma, P. (2004) An overview of the field of family business studies: current status and directions for future, Family Business Review, 17(1), 1-36.
Sharma, P. & Irving, G. (2005). Four Bases of Family Business Successor Commitment: Antecedents and Consequences. Entrepreneurship Theory and Practice 29 (1), 13-33
Sharma, P., & Manikutty, S. (2005). Strategic divestments in family firms: Role of family structure and community culture. Entrepreneurship Theory and Practice, 29, 293-312
Smith, B. F., & Amoako-Adu, B. (1999). Management succession and financial performance of family controlled firms. Journal of Corporate Finance 5, 341-368.
Sorenson, R. L. (1999). Conflict management strategies used in successful family businesses. Family Business Review, 12(4), 325-339
Stewart, & Hitt (2012). Why can't a family business be more like a nonfamily business? Modes of professionalization in family firms. Family Business Review,
Ucbasaran, D., Westhead, P., & Wright, M. (2001). The Focus of Entrepreneurship Research: Contextual and Process Issues. Entrepreneurship Theory & Practice, 25(4), 57-80.
Uzzi, B. (1997). Social structure and competition in interfirm networks: The paradox of embeddedness. Administrative Science Quarterly, 42, 35-67.
Venter, E., Boshoff, C, & Mass G. (2005). The Influence of successor-related factors on the succession process in small and medium-sized family businesses. Family Business Review, 18(4).
Ward, J.L. (1987). Keeping the family business healthy: How to plan for continuing growth. San Francisco: Jossey-Bass.
Zellweger, T. (2007). Time horizon, cost of equity capital, and generic investment strategies of firms. Family Business Review. 20 (1), 1-15.
Zellweger, T., Nason, R., & Nordqvist, M. (2012). From longevity to transgenerational value creation: The role of family entrepreneurial orientation. Family Business Review, forthcoming (published online)
Yu, A., Lumpkin, G. T., Sorenson, R. L., & Brigham, K. H. (2012). The Landscape of Family Business Outcomes A Summary and Numerical Taxonomy of Dependent Variables. Family Business Review, 25(1), 33-57.


Table 1: Descriptive Statistics and Correlations

Mean
S.D.
Min
Max
1
2
3
4
5
6
7
8
9
10
11
12
13
14
1. Outcome




1













2. EBIT (log)
11.53
0.06
9.84
12.96
0.02
1.00












3. Owner salary Mean (log)
12.62
0.48
7.47
14.75
-0.12
0.18
1.00











4. Leverage
4.90
25.17
-1839
796
-0.01
-0.02
0.01
1.00










5. Firm age
0.40
0.49
0
1
-0.02
0.08
0.04
-0.03
1.00









6. Owner mean age
48.28
7.89
17
84
0.07
0.04
0.01
-0.01
0.14
1.00








7. FT owners
0.92
0.20
0
1
-0.19
-0.04
0.48
0.02
0.04
-0.02
1.00







8. Non-mutual heirs
0.14
0.38
0
4
0.01
-0.01
0.01
0.00
-0.07
0.02
-0.01
1.00






9. Employees (log)
2.93
0.67
0
7.05
0.02
0.37
0.26
0.04
0.12
0.03
0.00
0.01
1.00





10. CEO exit
0.12
0.32
0
1
0.23
-0.02
-0.07
0.01
-0.01
0.00
-0.08
0.00
-0.01
1.00




11. Ownership disp.
2.81
2.00
1
51
0.00
0.06
0.01
-0.02
0.05
-0.10
-0.05
0.14
0.16
0.01
1.00



12. Young heirs (log)
0.86
0.70
0
4.09
-0.03
0.03
0.06
-0.01
0.02
-0.47
0.03
0.10
0.08
0.02
0.39
1.00


13. Adult heirs (log)
1.06
0.57
0
4.39
-0.01
0.06
0.01
0.01
0.15
0.43
-0.04
0.19
0.11
0.01
0.38
-0.17
1.00

14. Multiple gen.
0.34
0.47
0
1
0.05
0.06
-0.07
0.00
0.05
-0.08
-0.09
-0.02
0.09
0.02
0.24
0.16
0.29
1.00
15. Internal CEO
0.73
0.45
0
1
-0.06
-0.01
0.01
-0.01
0.00
0.07
0.02
0.00
-0.07
-0.11
-0.09
-0.07
-0.12
-0.29

Note: Correlations > "0.03" are statistically significant at p < 0.001


Table 2: Multinomial Logit Model: Internal Transfer Outcome Compared with the External Transfer Outcome
 
Model 1
Model 2

Base Model
SE
Full Model
SE
Control Variables




EBIT (log)
-0.182
(0.758)
-0.679
(1.029)
Owner salary mean (log)
-0.052
(0.110)
-0.109
(0.114)
Leverage
-0.002
(0.002)
-0.002
(0.002)
Firm age
0.127
(0.112)
0.004
(0.114)
Owner mean age
0.013*
(0.006)
0.041***
(0.008)
Full-time owners
0.066
(0.248)
0.224
(0.262)
Non-mutual heirs
0.122
(0.126)
0.144
(0.135)
Number of employees (log)
-0.101
(0.077)
-0.172*
(0.086)
CEO Exit
0.051
(0.127)
0.190
(0.134)
Hypotheses




H1: Ownership dispersion


0.154**
(0.046)
H2a: Young heirs (log)


0.055
(0.110)
H2b: Adult heirs (log)


-0.259*
(0.135)
H3: Multiple generations


1.919***
(0.130)
H4: Internal CEO


1.349***
(0.122)
_cons
3.040
(8.673)
6.715
(11.672)
Pseudo R2 (McFadden)
0.102

0.146

LR χ 2
1285.60***

1844.54***

Wald test
n/a

536.86***

Firm-year observations
12125



N
3829

 






Standard errors in in parentheses. All standard errors clustered on the firm level. Year and industry dummies included but not reported. Multinomial logit models tested against the external transfer outcome (comparison against continuation and shutdown are not reported in this table). * p < 0.05; ** p < 0.01; *** p < 0.001