Presentation and report on: The entrepreneurial orientation -performance relationship in private family firm

The entrepreneurial orientation–performance relationship in private family rms: the moderating role of socioemotional wealth Jelle Schepers •Wim Voordeckers • Tensie Steijvers •Eddy Laveren Accepted: 10 December 2013 / Published online: 22 December 2013 Springer Science+Business Media New York 2013 AbstractDrawing on socioemotional wealth (SEW) literature, this paper revisits the established entrepre- neurial orientation (EO)–performance relationship in a family business context. The main idea in entrepre- neurship literature is that EO leads to increased rm performance. We question this logic in a family business context because family related non- nancial goals, like SEW, may prevent the rm to reap the fruits of their entrepreneurial efforts. Speci cally, we argue that SEW engenders inef ciencies that place con- straints on the realization of the bene ts of entrepre- neurship. Therefore, we propose that a high level of SEW preservation hinders the transmission of the family rm’s EO into positive performance effects. To test this hypothesis, an empirical study was developed using a sample of 232 Belgian private family rms.

Robust linear regression analysis reveals that the positive effect of EO on nancial performance decreases as the level of SEW preservation increases.KeywordsEntrepreneurial orientation Private family rms Socioemotional wealth Firm performance JEL Classi cationsL21 L25 L26 1 Introduction For many years, researchers have argued that rms pursuing a high entrepreneurial orientation (i.e. a strategic posture that involves a propensity to be innovative, proactive and open to risk in exploring new products, services and markets [Covin and Slevin 1991]) perform better (e.g. Su et al.2011; Rauch et al.

2009; Wiklund and Shepherd2005; Covin and Slevin 1989). The implicit logic behind this pervasive belief seems to be that entrepreneurial rms will identify and pursue lucrative product/market opportunities which in turn will improve their company nancial perfor- mance (Zahra and Covin1995). Although this idea is widely accepted in the literature, empirical evidence showed that there exists considerable variation in the size and direction of reported relationships between entrepreneurial orientation (EO) and rm performance (Rauch et al.2009). These observations inspired researchers to apply a contingency framework incor- porating moderating variables that may explain vari- ations in the EO–performance relationship (Covin and Slevin1991; Lumpkin and Dess1996), ranging from J. Schepers W. Voordeckers (&) T. Steijvers Kizok Research Center, Hasselt University, Hasselt, Belgium e-mail: [email protected] J. Schepers e-mail: [email protected] E. Laveren University of Antwerp, Antwerp, Belgium 123 Small Bus Econ (2014) 43:39–55 DOI 10.1007/s11187-013-9533-5 external variables, such as environmental dynamism (e.g., Wiklund and Shepherd2005; Moreno and Casillas2008), to internal variables, such as strategic process variables (Covin et al.2006).

Recently, the EO–performance relationship has been investigated in a family business context. Family rms constitute an appropriate and unique context to analyze EO and its relationship with performance because of their distinctive set of ownership, manage- ment and governance conditions vis-a` -vis non-family rms (e.g. Casillas and Moreno2010; Huybrechts et al.2011). Moreover, the interaction between family and business has a signi cant impact on the decision- making process and entrepreneurial activities in family rms (Nordqvist et al.2008). However, family rms are not a homogenous group as different ‘‘types’’ of family rms seem to exist, based on differences in company ownership, management structures, and company objectives (Westhead and Howorth2007).

Therefore, several recent studies (e.g. Casillas and Moreno2010; Casillas et al.2010; Chirico et al.2011) investigated family rm speci c variables such as family and generational involvement as moderators on the EO–performance relationship. For example, Casi- llas et al. (2010) found that EO has a positive effect on rm growth in second-generation family rms. In addition, Casillas and Moreno (2010) reported that family involvement (in management and strategic decision-making processes) has a boosting effect on the relationship between the innovativeness dimension of EO and rm growth and a reducing effect on the relationship between the risk taking dimension of EO and rm growth.

Although these studies provided signi cant contri- butions to both the entrepreneurship and the family business literature, they are subject to a threat that is ubiquitous in family rm research. More speci cally, family business scholars often rely on reductionist proxies (e.g. percentage of family members in man- agement functions or composition of board of direc- tors) to gauge the degree of family in uence in these rms (Go´ mez-Mej ´ a et al.2011). Although these indicators are usually convenient, they only partly capture the essence of family rms (Chua et al.1999).

Recently, family rm scholars are attaching more and more importance to this essence approach and call for the inclusion of variables that relate to the noneco- nomic aspects and emotions of family businesses (Go´ mez-Mej ´ a et al.2011). Therefore, the aim of thispaper is to examine socioemotional wealth (SEW) preservation as a moderator on the EO–performance relationship. SEW refers to the non nancial aspects of the rm that meet the family’s affective needs such as identity, the ability to exercise family in uence, and the perpetuation of the family dynasty (Go´ mez-Mej ´ a et al.2007) and may drive family business behavior to a large extent. Because family rms are often loss averse when it comes to their SEW, they will behave in order to preserve these non- nancial bene ts which may have a signi cant effect on the EO–performance relationship. Recent literature has shown that SEW has two sides, a bright side and a dark side (Kellermanns et al.2012). This means that SEW concerns can lead to favorable (e.g. employee commitment, emotional attachment, better environmental performance) and unfavorable outcomes (e.g. altruism, incompetent family managers, inef cient use of rm resources) in family rms. In this paper we will argue that it is especially the dark side of SEW that moderates the relationship between EO and nancial performance.

While EO is known to be a resource-consuming strategic orientation (e.g. Covin and Slevin1991;Su et al.2011), it involves making large resource commitments in order to reach higher nancial performance. Hence, in private family rms, rm resources play a crucial role in the performance implications of EO. For that reason, we introduce SEW as a moderator on the EO–performance relationship because it provides insight into how family rms exploit their resources (Go´ mez-Mej ´ a et al.2011). Indeed, in private family rms, rm resources are often used inef ciently due to SEW considerations (Cruz et al.2012). By their own nature, family rms are characterized by a wide range of emotions and interpersonal linkages which may engender parental altruism (Schulze et al.

2003a) or managerial entrenchment (Gomez-Meija et al.2001). We draw on family business literature (e.g. Go´ mez-Mej ´ a et al.2011) to argue that family rms often face inef ciencies (like parental altruism and managerial entrenchment) as a result of their SEW preservation. For example, employment of incapable family members creates speci c agency costs (Lubatkin et al.2005; Cruz et al.2012) or may enlarge rent extraction in the family rm (Go´ mez- Mej ´ a et al.2011), leading to lower pro tability.

Consequently, from a nancial point of view, EO may increase the family rm’s sales level (top line 40J. Schepers et al.

123 in the pro t and loss account) but it does not automatically enhance their pro tability (bottom line in the pro t and loss account).

Our paper contributes to the literature in several ways. First, this study ts the call for incorporating moderating variables that potentially affect the rela- tionship between EO and performance (Covin and Slevin1991; Lumpkin and Dess1996). More in particular, we introduce SEW in the EO–performance debate as a family- rm-speci c variable that describes family rm’s behavior, rather than simply looking at the generational stage (e.g. Casillas et al.2010; Chirico et al.2011) that controls the rm or the proportion of family members involved in manage- ment functions (e.g. Casillas and Moreno2010). In doing so, we expand Covin and Slevin’s (1991) and Lumpkin and Dess’ (1996) contingency framework by introducing a new category of moderating variables, namely, behavioral moderators. Furthermore, this paper contributes to the family business literature because it introduces EO as a major determinant for a family rm’s nancial performance without ignoring the importance of non- nancial aspects in family businesses.

The remainder of this article is divided into ve sections. First, we explore the appropriate theoretical and empirical literature that relates to the EO– performance relationship. Second, SEW is introduced as a moderating variable in the EO–performance relationship and our central hypothesis is derived. In the third section, we elaborate our research method where we build on Brambor et al. (2006) and Kam and Franzese’s (2007) work to suggest that even if the coef cient of the interaction term is not signi cant, it is still possible that SEW may moderate the effect EO has on nancial performance. Fourth, the results of our empirical study will be presented and discussed.

Finally, the paper ends with a discussion section where the major conclusions are highlighted and future research paths are presented.

2 Theoretical and empirical background of the EO–performance relationship In almost 30 years of research, the phenomenon of an EO has become one of the major topics in the entrepreneurship literature. The concept of EO stems from Miller’s (1983) work, in which entrepreneurial rms are de ned as ‘‘those that are geared towards innovation in the product-market eld by carrying out risky initiatives, and which are the rst to develop innovations in a proactive way in an attempt to defeat their competitors’’ (p. 771). Although there have been various conceptions about EO’s components, research has converged on three core dimensions of EO (Miller and Le Breton-Miller2011): innovation, risk-taking, and proactiveness. Therefore, the current paper de nes EO as a rm-level construct where innovativeness, risk-taking, and proactiveness are assumed to covary, in line with Miller (1983) and Covin and Slevin’s (1991) conceptualization of EO. In this view, each organization falls somewhere along a conceptual continuum ranging from conservative (low EO) to entrepreneurial (high EO) (Covin and Slevin1988).

Although many different approaches and samples have been used, researchers generally agree on the fact that EO positively in uences rm performance. This widely accepted belief primarily stems from Scholl- hammer (1982, p. 210) who stated that ‘‘Entrepre- neurship is the key element for gaining competitive advantage and consequently greater nancial rewards.’’ However, Lumpkin and Dess (1996) were one of the rst scholars to discuss the EO–perfor- mance relationship by stressing the importance of viewing this relationship in a contingency framework.

Contingency theory suggests that certain key vari- ables, such as environmental or organizational vari- ables, need to be con gured to reach a t in order to obtain optimal performance. Current research aims at nding such key variables in order to explain additional performance variance. The possible exis- tence of variables that moderate the relationship between a rm’s entrepreneurial posture and rm performance was already recognized in 1991 by Covin and Slevin. They made a distinction between three classes of moderating variables, namely, internal, external, and strategic variables.

Several researchers are increasingly operationaliz- ing Covin and Slevin’s ( 1991) model, which indicates a tendency of incorporating moderating variables in EO–performance research. Resource availability (e.g.

Frank et al.2010; Moreno and Casillas2008; Wiklund and Shepherd2005) and environmental characteristics (e.g. Casillas et al.2010; Frank et al.2010; Van Doorn and Volberda2009; Moreno and Casillas2008; Wiklund and Shepherd2005; Lumpkin and Dess 2001) are by far the most widely used moderators in The moderating role of socioemotional wealth41 123 EO–performance studies. Several other studies inves- tigated additional moderating variables such as the stage of industry life cycle (Lumpkin and Dess2001), strategic process variables (Covin et al.2006), longevity (Runyan et al.2008), senior team attributes (Van Doorn and Volberda2009), and internal social exchange processes (De Clercq et al.2010). Within a family business context, generational involvement (Casillas et al.2010; Chirico et al.2011), and family involvement (Casillas and Moreno2010) have been studied as moderating variables.

Recently, Miller (2011) stressed that the issue of context may in uence EO and its relationship to performance. He stated that, ‘‘a good way of making context precise is to investigate a particular organiza- tion type’’ (Miller2011, p. 9). In this paper, we meet the needs of this call by investigating family busi- nesses as a particular organizational type because we believe the intersection between family business literature and entrepreneurship literature (e.g. Uhlaner et al.2012) has the potential to explain extra perfor- mance variance in family businesses. Family rms are the most dominant organizational form in the world (Dyer2003), but in the meantime they are so diverse that they cannot be treated as one single group of organizations (Westhead and Howorth2007). There- fore, instead of making generalized assumptions about their behavior, we add to current literature by directly measuring a deeper underlying variable that drives their behavior, namely, SEW. In the following section, we introduce SEW as a moderating variable on the EO–performance relationship in family businesses.

3 Socioemotional wealth as a moderator on the EO–performance relationship in family businesses 3.1 Socioemotional wealth Family rms are an important and prevalent type of rm that is often characterized by the family’s large undi- versi ed equity position and its control of leadership.

The interplay between the family and the business is often represented in Tagiuri and Davis’s (1996)three- circle model that makes a clear distinction between three subsystems, namely, the business-, the ownership-, and the family subsystem, each having its own goals and ambitions. In family rms, the business and the familyare often so intertwined that it is hard to distinguish where one ends and the other begins, with business goals often embraced by family goals (Sharma et al.1997).

When making business decisions, family rms combine a mix of family-oriented goals and business-oriented goals (Mahto et al.2010). Therefore, we introduce SEW, as an essential construct in family business literature, in order to analyze family business behavior.

Go´ mez-Mej ´ aetal.(2007, p. 106) de ne SEW as ‘‘non- nancial aspects of the rm that meet the family’s affective needs’’, and state that SEW preservation is often a goal in itself because family rms are loss averse when it comes to their SEW. Stated differently, family business owners, consciously or unconsciously, value non- nancial aspects that result from their family control such as identity, the perpetuation of the family dynasty, and the ability to exercise family in uence (Go´ mez-Mej ´ aetal.2007; Berrone et al.2010). To safeguard these non- nancial bene ts, family rms are willing to accept an increased risk of poor rm performance (Go´ mez-Mej ´ aetal.2007), as opposed to publicly traded rms where decisions are largely made based on nancial goals in order to maximize share- holder value (Mahto et al.2010). As a consequence, managerial decisions in family rms can be driven by a desire topreserveand enhance the family’s SEW because they are likely to see potential gains or losses in SEW as a key criterion in managing the rm (Go´ mez- Mej ´ aetal.2011; Berrone et al.2012).

One can argue that all types of rms may exhibit non- nancial goals (such as corporate social responsibility or customer satisfaction), but only family rms show signs of family-centered non nancial goals, which often relate to the family’s identity and reputation concerns (Zellweger et al.2010). Moreover, when the family member’s self-concept is strongly tied to the rm’s identity—where the rm often bears the person’s name—the individual derives considerable non-eco- nomic bene ts from membership in such an organiza- tion (Go´ mez-Mej ´ aetal.2007). According to social identity theory (e.g. Stets and Burke2000), people classify themselves and others into various social groups. Family membership is one of the most important social groups, and accordingly, the family business is directly tied to the family member identity. Therefore, family members will strive to portray a positive image of their rm in order to enhance their self-esteem and accomplish their need for acknowledgment and achievement (Mahto et al.2010). In line with this idea, 42J. Schepers et al.

123 family business owners are highly concerned with the rm’s reputation because they want the business to endure several family generations (Ward and Aronoff 1991), and to perpetuate the family dynasty. In addition, Zellweger and Astrachan (2008) state that family business owners subjectively value their ownership stake in monetary terms, indicating that the family’s perceived value of the rm may differ from the nancial value of their ownership stake and the private nancial bene ts of their control. Stated differently, the more importance the family attaches to the preservation of their SEW through their rm, the higher their perceived value of the rm will be. Again, these ndings are in line with the proposition of Go´ mez-Mej ´ aetal.(2007)that family rms are loss averse when it comes to decisions that affect their SEW because they are unwilling to sell the rm for only its nancial value.

3.2 The EO–performance relationship in family rms revisited In line with Miller (1983) and Covin and Slevin (1991), we see EO as the concurrent exhibition of innovativeness, proactiveness, and risk taking. A behavioral model of EO is suggested because behav- iors rather than attributes are what give meaning to the entrepreneurial process (Covin and Slevin1991; Covin and Lumpkin2011). Consequently, rms with a high degree of EO are characterized by a set of distinct but related behaviors that have the qualities of innovativeness, proactiveness, and risk taking. The primary tenet in entrepreneurship literature is that EO leads to improved performance (Lumpkin and Dess 1996). We question this basic supposition in a family business context. More speci cally, we question the logic that entrepreneurial activities (e.g. be the rst to introduce new products or services; dramatically change product or service lines) automatically enhance nancial performance since family rms often face inef ciencies as a result of their SEW preservation. Thus, the relationship between EO and increased nancial performance might be less straight- forward than often proposed.

As mentioned before, SEW is a relatively new concept which means that its relationship with nancial performance and EO remains to be studied (Berrone et al.2012). However, there are some studies that tried to link SEW to rm performance (e.g. Berrone et al.2010; Cruz et al.2012), but results are mixed. Also, theentrepreneurship literature has already linked some of the proposed SEW dimensions to entrepreneurial out- comes (e.g. Aldrich and Cliff2003; Lumpkin et al.

2010; Zahra et al.2004), but again with inconclusive results. When it comes to the EO–performance rela- tionship in private family rms, the extent to which SEW in uences this relationship has not yet been studied. In what follows, we illustrate how the ‘dark side’ of SEW (Kellermanns et al.2012)sti esthe transmission of EO into good nancial performance.

Since gains or losses in SEW represent the pivotal frame of reference that family rms use to make major decisions (Go´ mez-Mej ´ a et al.2007; Berrone et al.

2012), its impact on the EO–performance relationship cannot be underestimated. Fundamental to this theory is the notion that family rms frame problems in terms of assessing how actions will affect socioemotional endowment. When this endowment is threatened, the rm is willing to make decisions that are not driven by an economic logic (Berrone et al.2012). For example, family rms who place high importance on the preservation of their SEW tend to create or save jobs for family members in order to perpetuate the family dynasty (Go´ mez-Mej ´ a et al.2007 ). Furthermore, founder CEOs often have the possibility to be unusually generous to their children and relatives (Schulze et al.2001). Once the family has suf cient ownership for undisputed control, it can begin to free ride by exploiting the rm’s resources for personal bene ts and for privileges of family members (Schu- lze et al.2003a). Consequently, family employees are often given perquisites and privileges that they would not otherwise receive (Gersick et al.1997; Ward 1987). Even more, Kirchhoff and Kirchhoff (1987) found that when family member participation increases, wage and salary expenses increase as a percentage of revenue. So, our argument is that family rms are often saddled with additional costs as a result of their SEW preservation. Moreover, family rms which place high importance on maintaining family control and exercising family in uence, are often reluctant to professionalize (Go´ mez-Mej ´ a et al.

2011). That is, hiring outside managers and delegating authority are likely to reduce family control over strategic decisions (e.g. Go´ mez-Mej ´ a et al.2011).

Clearly, limiting executive management positions to family members (managerial entrenchment) can be problematic as the risk of employing low quality managers increases (e.g. Anderson and Reeb2003; The moderating role of socioemotional wealth43 123 Lubatkin et al.2005) which may compromise the pecuniary realization of entrepreneurship efforts.

Hence, we argue that even though a family rm is entrepreneurially oriented, the transmission of EO into good nancial performance can be hampered by SEW preservation because EO—as a resource consuming orientation—requires different resources in order to reach better nancial performance (e.g. Covin and Slevin1991;Suetal.2011). Free riding behavior, perquisites and privileges can lead to inef cient use of resources and additional costs (Cruz et al.2012), which prevent the family rm to translate EO into pro ts. More speci cally, entrepreneurial activity such as, for exam- ple, the introduction of new products, may increase the rm’s sales level (top line in the pro t and loss account) (e.g. Casillas and Moreno2010;Casillasetal.2010), but due to inef ciencies related to SEW preservation, this is not fully translated into higher pro ts (bottom line in the pro t and loss account). Thus, SEW has a price tag which constrains the family rm in realizing the bene ts of entrepreneurship and reaching higher pro tability levels. Therefore, we argue that SEW plays a crucial moderating role in the EO–performance relationship as SEW preservation has an impact on the mechanism that translates entrepreneurial efforts into pro ts. Thus, to understand how EO relates to nancial performance in a family business context, it is warranted to take into account the importance family members attach to the preservation of their SEW. Therefore we propose the following hypothesis:

HypothesisThe level of socioemotional wealth preservation will moderate the relationship between EO and a family rm’s nancial performance, in such a way that a family rm’s EO will have a less positive effect on nancial performance when the level of socioemotional wealth preservation increases.

4 Research method 4.1 Sample The sampling frame was taken in the 2002–2003 period 1 from a wider study investigating rmcharacteristics, strategic and environmental issues, board and management composition, succession, governance and performance issues in family busi- nesses in Flanders, which is the northern region of Belgium. In family business literature, there is a wide assortment of proxies that have been used in the empirical literature to de ne family rms (e.g. Ruth- erford et al.2008;Go´ mez-Mej ´ a et al.2011). In this study, we made use of commonly selected criteria of ownership and management control (Chua et al.1999) and CEO’s perception of being a family rm (West- head and Cowling1998) to select an operational de nition of family rms. As a consequence, in this paper, a rm is classi ed as a family rm if: (1) at least 50 % of the shares are owned by the family, and the family is responsible for the management of the company, or (2) at least 50 % of the shares are owned by the family, the company is not family-managed, but the CEO perceives the rm as a family rm. All family rms included in the sample were privately-owned. A total number of 3,400 rms were randomly selected from a family business database and a survey was mailed to the CEO. A response rate of 9.2 % resulted in 311 surveys, of which 295 were retained due to the deletion of non-family rms and incomplete cases. To run our own regression analysis, we deleted cases with missing values on relevant variables, resulting in a nal sample of 232 cases. Potential nonresponse bias was tested using two separate procedures. First, following the argument that late respondents are expected to be comparable to nonrespondents (Kanuk and Berenson1975), we differentiated between the 20 % earliest respondents and the 20 % latest respon- dents and performed severalttests and chi-square tests on the variables included in the analyses. The results revealed no signi cant differences on any of the variables, suggesting that there is no nonresponse bias in the results. Robustness checks with cut-off points at 10 and 30 % showed exactly the same results. In addition, performance indicators (dependent variable) were drawn from the Bel rst database of Bureau Van Dijk, covering 1.2 million Belgian companies. In Belgium, all rms are obliged to le their nancial statements to the National Bank of Belgium which is the primary source where Bureau Van Dijk retrieves its data from. We were able to match the data of our survey with the nancial records of the Bureau Van Dijk database using the company name. Furthermore, by using two different sources of data, common 1The sample was taken in the 2002–2003 period but our data is still appropriate to test our model since there is no indication to believe the EO–performance relationship is not stable over time (Rauch et al.2009).

44J. Schepers et al.

123 Table 1Means, standard errors, and correlations Variable Mean SE 1 2 3 4 5 6 7 8 9 10 11 1. Performance 14.3535 10.3248 1.0000 2. Maturity stage (dummy)0.4741 0.5004-0.0720 1.0000 3. Consolidation stage (dummy)0.1379 0.3456-0.0659-0.3798*** 1.0000 4. Firm age (Ln age) 3.3793 0.8645 0.0189 0.1329** 0.1864*** 1.0000 5. Firm size (Ln employees)2.6067 1.0619-0.0977 0.0825-0.0260 0.2407*** 1.0000 6. Manufacturing (dummy)0.3491 0.4777-0.0058-0.0436-0.0045 0.0972 0.2793*** 1.0000 7. Construction (dummy)0.1336 0.3410-0.0479-0.0431-0.0469-0.0258 0.1140*-0.2876*** 1.0000 8. Wholesale (dummy)0.2026 0.4028-0.0277 0.0368-0.0150-0.0725-0.1167*-0.3692***-0.1979*** 1.0000 9. Retail (dummy) 0.1552 0.3629-0.0991 0.0699-0.0333-0.0366-0.1955***-0.3139***-0.1683**-0.2160*** 1.0000 10. EO 24.1853 6.6087 0.1230*-0.0516-0.1269* 0.0452 0.2549*** 0.0192-0.0975 0.0885 0.0639 1.0000 11. SEW 14.8276 3.3349-0.0105-0.0105-0.0845-0.1379**-0.1606**-0.0082-0.0444-0.0190 0.0794 0.1038 1.0000 N=232 *,**,*** Signi cance at 0.10, 0.05 and 0.01, respectively The moderating role of socioemotional wealth45 123 method bias concerns are mitigated (Podsakoff et al.

2003). Sample characteristics of the data used in the regression analysis are speci ed in Table1.

4.2 Variables and measures Financial performanceis the most commonly used performance indicator when studying family busi- nesses (Rutherford et al.2008). Return on assets (ROA) was selected as the dependent variable because it is a well-understood and widely used accounting measure of nancial performance. We calculated each rm’s ROA as income before non cash items, interests and taxes divided by total assets as reported in the nancial statements, multiplied by 100. To fully capture the effect of EO on nancial performance, we measure performance in 2004 because it is often assumed that EO has a lagged effect on performance (Wiklund1999; Zahra1991).

Forentrepreneurial orientationwe use the nine- item scale validated by Miller/Covin and Slevin (1989) to gauge EO, capturing the rm’s innovation, proactiveness, and risk taking. Each individual sub- dimension includes three separate items. Recent studies have accentuated the need of consistency between the measurement model and the conceptual- ization of the EO construct (e.g. Covin and Wales 2011). Taking into account the consistency condition, we use the Miller/Covin and Slevin (1989) scale because this measure assesses combinations of EO’s elements via a composite dimension and is thus most consistent with our conceptualization of EO. Further- more, according to Covin and Wales (2011) and George (2011), the nine-item Miller/Covin and Slevin (1989) scale is the most commonly employed EO measure and has exhibited high levels of validity and reliability in numerous studies. In our study, the underlying EO dimensions were highly correlated and the alpha level for EO was found to be quite high (0.84). The correlation between the underlying EO dimensions allowed us to combine the three compo- nents and relate the composite EO-index to perfor- mance (Miller2011).

Socioemotional wealth preservation(SEW) can be de ned as ‘‘non- nancial aspects of the rm that meet the family’s affective needs’’ such as the ability to exercise family in uence, maintaining family control and the perpetuation of the family dynasty (Go´ mez- Mej ´ a et al.2007). In this study, SEW was measuredusing four questions taken from the Strategic Orien- tations of Small and Medium-Sized Enterprises (STRATOS) questionnaire (Bamberger1994, p. 399; Bamberger and Weir1990, p. 109): (1) maintaining family traditions/family character of the business, (2) creating/saving jobs for the family (both may be considered as proxies for the perpetuation of the family dynasty), (3) independence in ownership, and (4) independence in management (both may be considered as proxies for the ability to exercise family in uence and maintaining family control) (Go´ mez- Mej ´ a et al.2007; Goel et al.2013; Vandekerkhof et al.

forthcoming). The respondents were asked to indicate the importance they attach to each item on a 5-point Likert scale (1=totally unimportant, 5=very important). The rst item ‘‘maintaining family tradi- tions/family character of the business’’ refers to the role of affection and emotions in the family rm. SEW comes in a wide variety of related forms but the perpetuation of family values and traditions through the business is one important aspect of SEW (Go´ mez- Mej ´ a et al.2007). Hence, the family is loss averse when it comes to maintaining the family character of the business, even if this reduces ef cient exploitation (Cruz et al.2012). The second item ‘‘creating/saving jobs for the family’’ is related to ‘‘perpetuation of the family dynasty’’ from Go´ mez-Mej ´ a et al. (2007).

Also, creating or saving jobs for other family members is an essential part of the SEW construct because evidence has shown that passing the rm to the next generation (Zellweger et al.2012) and creating employment for family members (Cruz et al.2012) are both key goals for family rms. Since family rms are loss averse when it comes to handling their SEW (Go´ mez-Mej ´ a et al.2007), they will often limit vacancies to family members even if this con nes the size and the quality of the pool of potential employees.

The third item ‘‘independence in ownership’’ is also an important premise in the SEW debate and can be linked to Go´ mez-Mej ´ a et al.’s (2007) ‘‘family in u- ence’’. Zellweger et al. (2012, p. 851) state that:

‘‘Conceptually, the family’s control of the rm through ownership is critical to creating and preserv- ing socioemotional wealth since such control is what allows the family to pursue their interests through the rm. In other words, control is a necessary condition and plays a critical role in the theory of socioemotional wealth.’’ Independence in ownership by de nition implies that the family controls the rm and therefore 46J. Schepers et al.

123 it is highly relevant in measuring SEW. The fourth item ‘‘independence in management’’ refers to the family’s ability to exercise family in uence in the management of the rm. Family members can exert direct control over strategic decisions by appointing family members in the management team or selecting a family CEO. Basically, having the opportunity to be altruistic to other family members (e.g. providing management positions) adds to the family’s SEW (Go´ mez-Mej ´ a et al.2007; Schulze et al.2003b). The variables included in the scale are loaded on one single factor and capture the main elements of the SEW construct because they relate to the family’s affective bond and their psychological ownership over the business (Go´ mez-Mej ´ a et al.2007). The questions were summed into one single index, and the Cron- bach’s alpha reliability coef cient of the SEW scale was found to be 0.7, which is acceptable for explor- atory research (Hair et al.1998). Next, we validated our SEW measure using convergent validity and predictive validity. First, ‘convergent validity’ refers to the degree to which multiple endeavors to measure the same concept with different methods are in agreement (Venkatraman and Grant1986). Given that generation has been previously used as a proxy for SEW (Stockmans et al.2010), we look at the correlation between our measure of SEW and gener- ation by creating a dummy variable that equals zero if the family founder serves as CEO and 1 if a descendant serves as CEO. Here, we nd a negative correlation (-0.1696,p\0.01), which is in line with our expectations because SEW tends to decrease over generations (Go´ mez-Mej ´ a et al.2007). The negative correlation between SEW and generation provides evidence for convergent validity of our SEW measure.

Next, ‘predictive validity’ refers to the extent to which two measures that theoretically should be related, are in fact related (Venkatraman and Grant1986). In family business literature, SEW has been theoretically linked to (-) rm size (Go´ mez-Mej ´ a et al.2011; Go´ mez-Mej ´ a et al.2007), (-) rm age (Go´ mez-Mej ´ a et al.2007), (?) the proportion of family members on the board (Go´ mez-Mej ´ a et al.2011), and (?) the proportion of family members in the management team (Go´ mez-Mej ´ a et al.2011). Correlating these variables to our SEW measure provides the following results: rm size (-0.1606,p\0.05), rm age (-0.1379,p\0.05), proportion of family members on the board (?0.1672,p\0.01), and proportion offamily members in the management team (?0.3020, p\0.01). Again, these correlations are in line with theory, providing support for predictive validity of our SEW variable.

Control variables, similar to those in other EO– performance studies, were used to ensure proper model speci cation. In particular, we included numer- ous rm-level variables such as rm size (e.g. De Clercq et al.2010; Casillas et al.2010; Casillas and Moreno2010), measured as the natural logarithm of the number of full-time employees; rm age (e.g. De Clercq et al.2010; Casillas et al.2010; Casillas and Moreno2010), measured as the natural logarithm of the number of years the rm had been in business; rm industry (e.g. De Clercq et al.2010; Casillas et al.

2010; Casillas and Moreno2010), measured through four dummy variables that allow for ve major business lines to be differentiated: manufacturing, construction, wholesale, retail, and services; and the rm’s life-cycle stage, because it is stated that corporate life cycle may in uence the relationship between EO and performance (Miller and Le Breton- Miller2011). We created two dummy variables that allow for three main phases to be differentiated:

growth, maturity, and consolidation. This is a gener- ally accepted classi cation in the literature (Gray and Ariss1985) and simplifying a rm’s growth process into reduced categories is also standard practice (Phelps et al.2007). All control variables are derived from questions in the 2002–2003 survey or from the Bel rst database of Bureau Van Dijk.

5 Analysis and results We present the correlations and descriptive statistics for the variables in Table1, and supplementary descriptives are presented in Table2. In our sample, the mean value for a family rm’s EO, on a scale from 9 (low EO) to 45 (high EO), was found to be 24.2 with a standard deviation of 6.6. These ndings support the notion that the average family rm exhibits a moderate level of EO (e.g. Uhlaner et al.2012; Short et al.2009; Naldi et al.2007). In addition, the average rm included in our sample places relatively high impor- tance on the preservation of their SEW. On a scale from 4 (low SEW concerns) to 20 (the rm attaches high importance to the preservation of their SEW), a The moderating role of socioemotional wealth47 123 mean value of 14.8 was found with a standard deviation of 3.3. Also, the correlation table shows that SEW decreases with rm age and rm size, which is in line with prior literature (e.g. Go´ mez-Mej ´ a et al.

2007,2011). Besides this, an average rm in our sample is 41 years old and employs approximately 26 employees. A large amount of the rms are currently in the maturity stage of their life-cycle (47.4 %). In most cases, rms operate in the manufacturing (34.9 %) and the wholesale sector (20.3 %). Linear regression analysis was used to test our hypothesis.

In Table3, we provide the regression results for different models. The rst model, in which only control variables and the direct effect of EO on nancial performance were taken into account, pro- vides anR 2value of 0.0869 (p\0.05). Results indicate that EO is positively associated with a family rm’s nancial performance (b=0.22,p\0.05), which is in line with previous literature. The second model includes the interaction effect of SEW (EO*SEW) to test our central hypothesis. We use robust linear regression analysis to test this interaction effect, after mean centering EO and SEW to reduce multicollinearity concerns. At this point, the following regression model was used: nancial perfor- mance=a?b 1EO?b 2SEW?b 3EO*SEW?dcontrols?e.Here, we proposed that EO will have a less strong positive effect on nancial performance when the level of SEW increases. At rst glance, it appears that the results do not con rm our hypothesis because the interaction term is not signi cant.

There are however some important side notes that need to be taken into account. First, Brambor et al.

(2006; p. 70) note that ‘‘The coef cients in interaction models no longer indicate the average effect of a variable as they do in an additive model. Even more important to remember is that the analyst is not directly interested in the signi cance or insigni cance of the model parameters per se anyway. Instead, the analyst who employs a multiplicative interaction model is typically interested in the marginal effect of X on Y’’. The marginal effect of X (EO) on Y ( nancial performance) can be expressed by the following equation: ofinancial performance o EO¼b 1þb 3SEW.

Therefore, we look at the marginal effect of X (EO) on Y ( nancial performance) by taking into account the relevant elements of the variance–covariance matrix and recalculate the standard errors (Brambor et al.

2006). In doing so, SEW can turn out to have a signi cant moderating impact on the relationship between EO and performance for a certain range of Table 2Supplementary descriptives Descriptive Minimum Maximum Mean Standard error Entrepreneurial orientation (9–45) 9 42 24.18534 6.608741 Socioemotional wealth (4–20) 4 20 14.82759 3.334915 Financial performance-13 72 14.35345 10.32482 Number of employees 1 372 25.93533 43.71006 Firm age 3 362 40.68103 39.2241 DescriptivePercentage of observations The rm’s life-cycle stage Growth stage38.8 Maturity stage47.4 Consolidation stage13.8 Firm industry Manufacturing34.9 Construction13.3 Wholesale20.3 Retail15.5 Services16.0 48J. Schepers et al.

123 values even though Table3reveals an insigni cant moderating effect. 2The solid line in Fig.1presents the marginal effect of EO on nancial performance.

The dotted lines surrounding the solid line present the 95 % con dence interval, which allows us to deter- mine the conditions under which EO has a statistically signi cant effect on nancial performance. Thus, the EO–performance relationship is signi cant when both the upper and lower bounds of the con dence interval are above (or below) the zero line. The gure shows that a rm’s EO has a signi cant positive effect on a rm’s nancial performance when the level of SEW is situated between 12 and 17. Looking at our sample, we see that 59.1 % of the family rms are characterizedby a level of SEW situated in this range. Within this interval, the positive effect declines as the level of SEW increases. For extremely large (18–20) values of SEW, it seems that EO no longer affects performance.

This means that when the family attaches too much importance on the preservation of their SEW, the dark side of SEW becomes too dominant and prevents that entrepreneurial efforts are successfully converted into good nancial performance, which is exactly in line with our hypothesis. Since 25.4 % of the family rms in our sample are characterized by SEW values between 18 and 20, our hypothesis is supported by almost 85 % of our data. Only a small proportion of our sample, namely 15.5 %, is characterized by extremely small SEW values (4–11). Here, we nd no support for our hypothesis which is probably due to the rather substantial range of values for which we only have very few cases available. Table 3Linear regression analysis: three models Variable Model 1 Model 2 Model 3 bSEbSEbSE Controls Life-cycle stage a Maturity stage-2.0271 1.6502-2.0484 1.6517-1.8834 1.5033 Consolidation stage-3.6540 2.5435-3.5820 2.5574-3.6368 2.1953 Firm characteristcs Firm age 0.8890 1.0096 0.9027 1.0368 0.7409 0.8330 Firm size-1.4765** 0.7387-1.4523** 0.7141-1.5402** 0.7188 Industry b Manufacturing-3.7206 2.6491-3.7597 2.6833-3.7356* 2.0735 Construction-4.6804 3.3587-4.6583 3.3633-4.8401* 2.5326 Wholesale-5.0219* 2.5888-4.9968* 2.5910-5.1173** 2.2216 Retail-6.9979** 2.7372-6.9413** 2.7394-6.8271** 2.3674 Hypothesis EO 0.2203** 0.0944 0.2180** 0.0960 SEW 0.0442 0.2061 EO*SEW-0.0134 0.0306 EO*SEWhigh0.1765 0.1125 EO*(1-SEWhigh)0.2561** 0.1108 R 2 0.0869 0.0879 0.0940 F1.99** 1.61* 2.29** N=232 *,**,*** Signi cance at 0.10, 0.05 and 0.01, respectively aSuppressed category for the rm’s life-cycle stage is ‘‘growth stage’’bSuppressed category for the rm’s industry is ‘‘services’’ 2In general, researchers should always calculate marginal effects even if the interaction coef cient is signi cant (Brambor et al.2006; Kam and Franzese2007).

The moderating role of socioemotional wealth49 123 To con rm our central hypothesis, we performed a more straightforward post hoc regression analysis.

Namely, we created a dummy variable ‘SEWhigh’ that equals one when the level of SEW is situated between the range of 18 and 20. This cut-off value follows from our Brambor analysis and coincides with the SEW values where the marginal effect of EO on rm performance becomes insigni cant. At this point, the following estimated regression model was used (Yip and Tsang2007): nancial performance=a?b 1 EO*SEWhigh?b 2 EO*(1-SEWhigh)?dcon- trols?e. The results from model 3 in Table3indicate that EO has a signi cant positive effect (b=0.2561, p\0.05) on nancial performance when SEW is lower than 18 (EO*[1-SEWhigh]). On the other hand, when SEW is higher than 18, EO no longer affects nancial performance (EO*SEWhigh). These results are in line with our previous ndings. Taken together, our results indicate that the marginal effect of EO on rm performance decreases when the family attaches more importance to the preservation of their SEW which provides support for our hypothesis. The moderating role of SEW becomes even more domi- nant for extremely large values of SEW because our results indicate that higher EO is no longer translated in better nancial performance when the family attaches too much importance on the preservation of their SEW.

Although most EO–performance studies have found a positive linear relationship between EO and performance, there is some empirical evidence that aninverse curvilinear relationship may exist due to differences in market context (e.g. Tang et al.2008).

In our study, a post hoc analysis supported a linear relationship between EO and performance. The inclusion of EO-square in our regression model did not change our results, suggesting that there is no curvilinear relationship between EO and performance.

6 Discussion and conclusion EO is a corollary concept that emerged primarily from the strategic management literature (Lumpkin and Dess1996) while SEW can be seen as a ‘‘homegrown’’ (Berrone et al.2010, p. 2) theoretical formulation within the family business eld. Nevertheless, our study shows that both concepts in uence the behavior of private family rms and ultimately helps explain performance variance in these rms. Indeed, prior literature shows that EO must be seen as a resource consuming strategic orientation (e.g. Covin and Slevin 1991) since it requires different types of resources in order to reach better nancial performance. For example, access to nancial resources (e.g. Frank et al.2010; Wiklund and Shepherd2005) and physical resources (Moreno and Casillas2008) are found to be important moderating variables in the EO–perfor- mance relationship. SEW, as a dominant paradigm in the family business eld, provides insight into how family rms exploit their resources (Go´ mez-Mej ´ a et al.2011) and consequently adds to our Fig. 1Marginal effect of EO on a rm’s nancial performance as SEW changes. Percentage distribution of the SEW variable: SEW values 4–11=15.5 % of the sample; SEW values 12–17=59.1 % of the sample; SEW values 18–20=25.4 % of the sample 50J. Schepers et al.

123 understanding of the EO–performance relationship in private family rms.

Results show that EO positively in uences nan- cial performance. These ndings are in line with previous research (e.g. De Clercq et al.2010; Rauch et al.2009; Wiklund and Shepherd2005; Covin and Slevin1989) and indicate that the positive EO– performance link also applies to a family business context. This demonstrates the importance of having an EO in family rms because it helps to create or sustain a higher level of nancial performance.

Consistent with our central hypothesis, the level of a rm’s EO is less positively related to rm performance when the level of SEW preservation is high and even becomes insigni cant when SEW is extremely high.

These ndings are in line with our central reasoning and contribute to the literature on the dark side of SEW (Kellermanns et al.2012). We demonstrated how SEW hampers the transmission of EO into nancial performance gains. For rms with extremely high SEW values, it seems that a higher entrepreneurial orientation does not affect nancial performance.

These ndings can be interpreted as an extension of our central reasoning. Namely, it might mean that when family rms placetoo muchimportance on the preservation of their SEW, rm resources will be used inef ciently (Cruz et al.2012). Thus, even if the rm demonstrates high EO, they won’t be able to reap the nancial bene ts.

It is straightforward to see what happens in a pro t and loss account. For example, the introduction of a new product can increase the rm’s sales level (top line), but when the family places too much importance on the preservation of their SEW, excessive costs (e.g.

free riding costs, perquisites) will prevent the rm from translating entrepreneurial efforts into higher pro t gures (bottom line). Especially in entrepre- neurship literature, there has been a tendency to equate sales growth with business success. However, sales growth is only one step towards business success (pro tability) because other factors, like SEW, need to be taken into account. Therefore, while EO is essen- tially a growth orientation (Lumpkin and Dess1996), which is almost universally portrayed as a good thing, we believe it does notautomaticallymake the rm pro table. These ndings seem to be tantamount to an ongoing discussion in growth literature where growth without pro tability is not always seen as a sign of sound development (Davidsson et al.2009).Next, while the importance of the EO–performance relationship has been frequently recognized, only a small number of scholars has discussed this phenom- enon in a family business context (e.g. Cruz and Nordqvist2012; Zellweger and Sieger2012), which is rather distinct from other organizational contexts due to the combination of family and business systems.

Thus, family rm speci c variables that may affect the EO–performance relationship remain largely unex- plored. We examine family rm behavior and its impact on the EO–performance relationship by taking into account a variable that drives their behavior, namely, SEW. In doing so, we aim to contribute to the literature in at least two ways. First, by theoretically and empirically investigating the repercussions of family rm behavior on the EO–performance rela- tionship, we illustrate how SEW preservation changes the effect EO has on a rm’s nancial performance without ignoring the heterogeneous character of family rms. Generally speaking, we develop Covin and Slevin’s (1991) and Lumpkin and Dess’ (1996) works by introducing a behavioral variable into their contingency framework. This behavioral approach is a relatively new attempt to shed light on the black box between EO and performance and is not limited to the family business context only. Behavioral moderators, like SEW, have the potential to explain additional performance variance in all kinds of rms, over and above the traditional environmental and organiza- tional contingencies, because they will certainly in uence the implementation of a rm’s EO. We believe our results are inspiring and hope future research will build on our train of thought.

Furthermore, we want to point out that our results can be interpreted not only in light of EO–performance studies, but also are complementary to other research elds. More speci c, disentangling the black box between an input variable and an output variable is attracting increased attention from researchers of different research elds (e.g. Fey et al.2009; Daily et al.2003). In this study, we contribute to the attempt to partially open up the black box of the EO– performance link in family rms. Investigating this black box is especially relevant in the context of private family rms because they are often seen as complex systems (Tagiuri and Davis1996) that continuously need to con gure certain variables in order to be successful (Miller and Le Breton-Miller 2006). Thus, simply having an EO is not a suf cient The moderating role of socioemotional wealth51 123 condition for a family business to be successful because the transmission of EO into higher nancial performance can be hampered by certain (family related) variables. We contributed to the literature in partially opening up the black box by means of presenting SEW as a variable that impedes the transmission of EO into good nancial performance.

Moreover, our ndings are complementary to other recent arguments in the family rm literature such as family rm heterogeneity. In particular, family busi- ness literature can be divided into two major streams of literature that describe the diversity in family rms (Sharma and Nordqvist2007). The rst stream can be described as the ‘components of family involvement’ (e.g. Klein et al.2005; Westhead and Cowling1998), which simply captures the extent and mode of family involvement in ownership, management, governance, and succession. The second stream, known as the ‘essence approach’ (Chua et al.1999), focuses on the repercussions of family involvement on the behavior and decision making styles in these rms. In this context, our paper adds to the essence approach in a way that it underlines the giant diversity in family rms by accentuating an underlying variable that drives their behavior. Although we did not explicitly hypothesize, our ndings reveal that family rms attach varying importance to the preservation of their SEW, which seems to determine their behavior. This suggests that simply de ning a family business by its components does not necessarily capture its essence, because the variability in SEW preservation entails differences in their actual behavior. Indeed, some rms are highly attached to the preservation of their SEW whereas others attach rather moderate impor- tance to the preservation of these noneconomic factors. Nevertheless, it is acknowledged that SEW plays a pivotal role in the behavior of family rms (Go´ mez-Mej ´ a et al.2011), but the way in which it in uences organization outcomes is a highly desired future research area since SEW has two sides, a dark side and a bright side (Kellermanns et al.2012).

Depending on the research model, we encourage scholars to re ect on this ambidextrous feature of SEW and decide whichever side of SEW is dominant in the context of their study. For example, SEW can increase the commitment and trustworthiness of family employees (bright side) but at the same time SEW may lead to inef cient use of rm resources and incompetent family management (dark side). Webelieve the concept of SEW can only prosper if researchers bear in mind its two faces because a uni ed view of SEW might be too straightforward.

Our study also has some limitations which may provide interesting avenues for future research. First, it would be interesting to expand the model by incorporating other moderating variables that help to gain a more profound understanding of family busi- ness performance. In particular, other variables relat- ing to family rm behavior, such as family orientation (Lumpkin et al.2008) or long-term orientation (Le Breton-Miller and Miller2006), can be used as supplementary variables to build more complex mod- els which might improve researcher’s knowledge concerning the family rm’s EO–performance rela- tionship. Second, our work has used the Miller/Covin and Slevin (1989) scale to capture a rm’s EO.

Although this scale is the most commonly employed EO measure in the literature (Covin and Wales2011; George2011; George and Marino2011), other scales exist which may provide interesting additional insights. Similar reasoning applies for the SEW scale where more re nements can be made in future research.

Although there is no indication to believe the EO– performance relationship is not stable over time, it might be an interesting avenue for future research to replicate our study with data from a period of economic crisis, such as experienced in recent years, to test whether these extreme circumstances affect the relationship between EO and performance in private family rms. AcknowledgmentsThis research was partially funded by the National Bank of Belgium.

References Aldrich, H. E., & Cliff, J. E. (2003). The pervasive effects of family on entrepreneurship: Toward a family embedded- ness perspective.Journal of Business Venturing, 18(5), 573–596.

Anderson, R. C., & Reeb, D. M. (2003). Founding-family ownership and rm performance: Evidence from the S&P 500.Journal of Finance, 58(3), 1301–1328.

Bamberger, I. (1994).Product/market strategies of small and medium-sized enterprises. Aldershot, Hants: Ashgate Publishing Group.

Bamberger, I., & Weir, A. (1990).Strategic orientations of small European businesses: The STRATOS group. Alder- shot, Hants: Gower Publishing Company. 52J. Schepers et al.

123 Berrone, P., Cruz, C. C., & Go´ mez-Mej ´ a, L. R. (2012). So- cioemotional wealth in family rms: Theoretical dimen- sions, assessment approaches, and agenda for future research.Family Business Review, 25(3), 258–279.

Berrone, P., Cruz, C. C., Go´ mez-Mej ´ a, L. R., & Larraza Kin- tana, M. (2010). Socioemotional wealth and corporate response to institutional pressures: Do family-controlled rms pollute less.Administrative Science Quarterly, 55(1), 82–113.

Brambor, T. C., Clark, W. R., & Golder, M. (2006). Under- standing interaction models: Improving empirical analy- ses.Political Analysis, 14(1), 63–82.

Casillas, J. C., & Moreno, A. M. (2010). The relationship between entrepreneurial orientation and growth: The moderating role of family involvement.Entrepreneurship and Regional Development, 22(3–4), 265–291.

Casillas, J. C., Moreno, A. M., & Barbero, J. L. (2010). A con gurational approach of the relationship between entrepreneurial orientation and growth of family rms.

Family Business Review, 23(1), 27–44.

Chirico, F., Sirmon, D. G., Sciascia, S., & Mazzola, F. (2011).

Resource orchestration in family rms: Investigating how entrepreneurial orientation, generational involvement, and participative strategy affect performance.Strategic Entre- preneurship Journal, 5(4), 307–326.

Chua, J. H., Chrisman, J. J., & Sharma, P. (1999). De ning the family business by behavior.Entrepreneurship Theory and Practice, 23(4), 19–39.

Covin, J. G., Green, K. M., & Slevin, D. P. (2006). Strategic process effects on the entrepreneurial orientation–sales growth rate relationship.Entrepreneurship Theory and Practice, 30(1), 57–81.

Covin, J. G., & Lumpkin, G. T. (2011). Entrepreneurial orien- tation theory and research: Re ections on a needed con- struct.Entrepreneurship: Theory & Practice,35(5), 855–872.

Covin, J. G., & Slevin, D. P. (1988). The in uence of organi- zation structure on the utility of an entrepreneurial top management style.Journal of Management Studies, 25(3), 217–234.

Covin, J. G., & Slevin, D. P. (1989). Strategic management of small rms in hostile and benign environments.Strategic Management Journal, 10(1), 75–87.

Covin, J. G., & Slevin, D. P. (1991). A conceptual model of entrepreneurship as rm behavior.Entrepreneurship The- ory and Practice, 16(1), 7–24.

Covin, J. G., & Wales, J. W. (2011). The measurement of entrepreneurial orientation.Entrepreneurship Theory and Practice, 36(4), 677–702.

Cruz, C., Justo, R., & De Castro, J. (2012). Does family employment enhance MSEs performance? Integrating so- cioemotional wealth and family embeddedness perspec- tives.Journal of Business Venturing, 27(1), 62–76.

Cruz, C., & Nordqvist, M. (2012). Entrepreneurial orientation in family rms: A generational perspective.Small Business Economics, 38(1), 33–49.

Daily, C. M., Dalton, D. R., & Cannella, A. A. (2003). Corporate governance: Decades of dialogue and data.Academy of Management Review, 28(3), 371–382.

Davidsson, P., Steffens, P., & Fitzsimmons, J. (2009). Growing pro table or growing from pro ts: Putting the horse infront of the cart?Journal of Business Venturing,24(4), 388–406.

De Clercq, D., Dimov, D., & Thongpapanl, N. (2010). The moderating impact of internal social exchange processes on the entrepreneurial orientation–performance relation- ship.Journal of Business Venturing, 25(1), 87–103.

Dyer, W. G. (2003). The family: The missing variable in orga- nizational research.Entrepreneurship Theory and Prac- tice, 27(4), 401.

Fey, C. F., Morgulis-Yakushev, S., Hyeon, J. P., & Bjo¨ rkman, I.

(2009). Opening the black box of the relationship between HRM practices and rm performance: A comparison of MNE subsidiaries in the USA, Finland, and Russia.Jour- nal of International Business Studies, 40(4), 690–712.

Frank, H., Kessler, A., & Fink, M. (2010). Entrepreneurial ori- entation and business performance—A replication study.

Schmalenbach Business Review, 62(2), 175–198.

George, B. A. (2011). Entrepreneurial orientation: An exami- nation of the consequences of differing construct repre- sentations.Journal of Management Studies, 48(6), 1291–1313.

George, B. A., & Marino, L. (2011). The epistemology of entrepreneurial orientation: Conceptual formation, mod- eling, and operationalization.Entrepreneurship Theory and Practice, 35(5), 989–1024.

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.

Goel, S., Voordeckers, W., Van Gils, A., & van den Heuvel, J.

(2013). CEO’s empathy and salience of socioemotional wealth in family SMEs—The moderating role of external directors.Entrepreneurship and Regional Development, 25(3/4), 111–134.

Go´ mez-Mej ´ a, L. R., Cruz, C., Berrone, P., & De Castro, J.

(2011). The bind that ties: Socioemotional wealth preser- vation in family rms.The Academy of Management Annals, 5(1), 653–707.

Go´ mez-Mej ´ a, L. R., Haynes, K., Nu´ n˜ ez-Nickel, M., Jacobson, K., & Moyano-Fuentes, J. (2007). Socioemotional wealth and business risks in family-controlled rms: Evidence from Spanish olive oil mills.Administrative Science Quarterly, 52(1), 106–137.

Gomez-Mejia, L. R., Nun˜ ez-Nickel, M., & Gutierrez, I. (2001).

The role of family ties in agency contracts.Academy of Management Journal,44(1), 81–95.

Gray, B., & Ariss, S. (1985). Politics and strategic change across organizational life cycles.Academy of Management Review, 10(4), 707–723.

Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C.

(1998).Multivariate data analysis. Upper Saddle River, NJ: Prentice Hall.

Huybrechts, J., Voordeckers, W., Lybaert, N., & Vandemaele, S. (2011). The distinctiveness of family- rm intangibles: A review and suggestions for future research.Journal of Management & Organization,17(2), 268–287 Kam, C. D., & Franzese, R. J. (2007).Modeling and interpreting interactive hypotheses in regression analysis. Michigan:

The University of Michigan Press.

Kanuk, L., & Berenson, C. (1975). Mail surveys and response rates: A literature review.Journal of Marketing Research, 12(4), 440–453. The moderating role of socioemotional wealth53 123 Kellermanns, F. W., Eddleston, K. A., & Zellweger, T. M.

(2012). Extending the socioemotional wealth perspective:

A look at the dark side.Entrepreneurship Theory and Practice, 36(6), 1175–1182.

Kirchhoff, B. A., & Kirchhoff, J. J. (1987). Family contributions to productivity and pro tability in small businesses.

Journal of Small Business Management, 25(4), 25–31.

Klein, S. B., Astrachan, J. H., & Smyrnios, K. X. (2005). The F-PEC scale of family in uence: Construction, validation, and further implication for theory.Entrepreneurship The- ory and Practice, 29(3), 321–339.

Le Breton-Miller, I., & Miller, D. (2006). Why do some family businesses out-compete? Governance, long-term orienta- tions, and sustainable capability.Entrepreneurship Theory and Practice, 30(6), 731–746.

Lubatkin, M., Schulze, W., Ling, Y., & Dino, R. (2005). The effects of parental altruism on the governance of family- managed rms.Journal of Organizational Behavior, 26(3), 313–330.

Lumpkin, G. T., Brigham, K. H., & Moss, T. W. (2010). Long- term orientation: Implications for the entrepreneurial ori- entation and performance of family businesses.Entrepre- neurship and Regional Development, 22(3/4), 241–264.

Lumpkin, G. T., & Dess, G. G. (1996). Clarifying the entre- preneurial orientation construct and linking it to perfor- mance.Academy of Management Review, 21(1), 135–172.

Lumpkin, G. T., & Dess, G. G. (2001). Linking two dimensions of entrepreneurial orientation to rm performance: The moderating role of environment and industry life cycle.

Journal of Business Venturing, 16(5), 429–451.

Lumpkin, G. T., Martin, W., & Vaughn, M. (2008). Family orientation: Individual-level in uences on family rm outcomes.Family Business Review, 21(2), 127–138.

Mahto, R. V., Davis, P. S., Pearce, J. A., I. I., & Robinson, R. B., Jr. (2010). Satisfaction with rm performance in family businesses.Entrepreneurship Theory and Practice, 34(5), 985–1001.

Miller, D. (1983). The correlates of entrepreneurship in three types of rms.Management Science, 29(7), 770–790.

Miller, D. (2011). Miller 1983. Revisited: A re ection on EO research and some suggestions for the future.Entrepre- neurship Theory and Practice, 35(5), 873–894.

Miller, D., & Le Breton-Miller, I. (2006). Priorities, practices and strategies in successful and failing family businesses:

An elaboration and test of the con guration perspective.

Strategic Organization, 4(4), 379–407.

Miller, D., & Le Breton-Miller, I. (2011). Governance, social identity, and entrepreneurial orientation in closely held public companies.Entrepreneurship Theory and Practice, 35(5), 1051–1076.

Moreno, A. M., & Casillas, J. C. (2008). Entrepreneurial ori- entation and growth of SMEs: A causal model.Entrepre- neurship Theory and Practice, 32(3), 507–528.

Naldi, L., Nordqvist, M., Sjo¨ berg, K., & Wiklund, J. (2007).

Entrepreneurial orientation, risk taking, and performance in family rms.Family Business Review, 20(1), 33–47.

Nordqvist, M., Habbershon, T. G., & Melin, L. (2008). Trans- generational entrepreneurship: Exploring entrepreneurial orientation in family rms. In H. Landstro¨ m, D. Smallbone, H. Crijns, & E. Laveren (Eds.),Entrepreneurship, sus- tainable growth and performance: Frontiers in Europeanentrepreneurship research(pp. 93–116). London, UK:

Edward Elgar.

Phelps, R., Adams, R., & Bessant, J. (2007). Life cycles of growing organizations: A review with implications for knowledge and learning.International Journal of Man- agement Reviews, 9(1), 1–30.

Podsakoff, P. M., MacKenzie, S. B., Jeong-Yeon, L., & Podsakoff, N. P. (2003). Common method biases in behavioral research:

A critical review of the literature and recommended reme- dies.Journal of Applied Psychology, 88(5), 879–903. Rauch, A., Wiklund, J., Lumpkin, G. T., & Frese, M. (2009).

Entrepreneurial orientation and business performance: An assessment of past research and suggestions for the future.

Entrepreneurship Theory and Practice, 3(3), 761–787.

Runyan, R., Droge, C., & Swinney, J. (2008). Entrepreneurial orientation versus small business orientation: What are their relationships to rm performance?Journal of Small Business Management, 46(4), 567–588.

Rutherford, M. W., Kuratko, D. F., & Holt, D. T. (2008).

Examining the link between ‘‘familiness’’ and perfor- mance: Can the F-PEC untangle the family business theory jungle.Entrepreneurship Theory and Practice, 32(6), 1089–1109.

Schollhammer, H. (1982). Internal corporate entrepreneurship.

In C. A. Kent, D. L. Sexton, & K. H. Vesper (Eds.), Encyclopedia of entrepreneurship(pp. 209–223). Engle- wood Cliffs, NJ: Prentice Hall.

Schulze, W. S., Lubatkin, M. H., & Dino, R. N. (2003a).

Exploring the agency consequences of ownership disper- sion among the directors of private family rm.Academy of Management Journal, 46(2), 179–194.

Schulze, W. S., Lubatkin, M. H., & Dino, R. N. (2003b). Toward a theory of agency and altruism in family rms.Journal of Business Venturing, 18(4), 473.

Schulze, W. S., Lubatkin, M. H., Dino, R. N., & Buchholtz, A.

K. (2001). Agency relationships in family rms: Theory and evidence.Organization Science, 12(2), 99–116.

Sharma, P., Chrisman, J. J., & Chua, J. H. (1997). Strategic management of the family business: Past research and future challenge.Family Business Review, 10(1), 1–35.

Sharma, P., & Nordqvist, M. (2007). A typology for capturing the heterogeneity of family rms.Academy of Management Annual Meeting Proceedings,1–6. doi:10.5465/AMBPP.

2007.26501993.

Short, J. C., Payne, G. T., Brigham, K. H., Lumpkin, G. T., & Broberg, J. C. (2009). Family rms and entrepreneurial orientation in publicly traded rms: A comparative ana- lysis of the S&P 500.Family Business Review, 22(1), 9–24.

Stets, J. E., & Burke, P. J. (2000). Identity theory and social identity theory.Social Psychology Quarterly, 63(3), 224–237.

Stockmans, A., Lybaert, N., & Voordeckers, W. (2010). So- cioemotional wealth and earnings management in private family rms.Family Business Review, 23(3), 280–294.

Su, Z., Xie, E., & Li, Y. (2011). Entrepreneurial orientation and rm performance in new ventures and established rms.

Journal of Small Business Management, 49(4), 558–577.

Tagiuri, R., & Davis, J. (1996). Bivalent attributes of the family rm.Family Business Review, 9(2), 199–208.

Tang, J., Tang, Z., Marino, L. D., Zhang, Y., & Li, Q. (2008).

Exploring an inverted U-shape relationship between 54J. Schepers et al.

123 entrepreneurial orientation and performance in Chinese ventures.Entrepreneurship Theory and Practice, 32(1), 219–239.

Uhlaner, L. M., Kellermanns, F. W., Eddleston, K. A., & Hoy, F.

(2012). The entrepreneuring family: A new paradigm for family business research.Small Business Economics, 38(1), 1–11.

Van Doorn, S., & Volberda, H. (2009). Entrepreneurial orien- tation and rm performance: The role of the senior team.

Academy of Management Annual Meeting Proceedings, 1–6. doi:10.5465/AMBPP.2009.44257958.

Vandekerkhof, P., Steijvers, T., Hendriks, W., & Voordeckers, W. (forthcoming). The effect of organizational character- istics on the appointment of nonfamily managers in private family rms: The moderating role of socioemotional wealth.Family Business Review.

Venkatraman, N., & Grant, J. H. (1986). Construct measurement in organizational strategy research: A critique and pro- posal.Academy of Management Review, 11(1), 71–87.

Ward, J. L. (1987).Keeping the family business healthy. San Francisco: Jossey-Bass.

Ward, J. L., & Aronoff, C. E. (1991). Trust gives you the advantage.Nation’s Business, 79(8), 42–45.

Westhead, P., & Cowling, M. (1998). Family rm research: The need for a methodological rethink.Entrepreneurship Theory and Practice, 23(1), 31–56.

Westhead, P., & Howorth, C. (2007). Types of private family rms:

An exploratory conceptual and empirical analysis.Entre- preneurship and Regional Development, 19(5), 405–431.

Wiklund, J. (1999). The sustainability of the entrepreneurial orientation–performance relationship.Entrepreneurship Theory and Practice, 24(1), 37–48.Wiklund, J., & Shepherd, D. (2005). Entrepreneurial orientation and small business performance: A con gurational approach.Journal of Business Venturing, 20(1), 71–91.

Yip, P. S. L., & Tsang, E. W. K. (2007). Interpreting dummy variables and their interaction effects in strategy research.

Strategic Organization, 5(1), 13–30.

Zahra, S. A. (1991). Predictors and nancial outcomes of cor- porate entrepreneurship: An exploratory study.Journal of Business Venturing, 6(4), 258–282.

Zahra, S. A., & Covin, J. G. (1995). Contextual in uences on the corporate entrepreneurship–performance relationship: A longitudinal analysis.Journal of Business Venturing, 10(1), 43–58.

Zahra, S. A., Hayton, J. C., & Salvato, C. (2004). Entrepre- neurship in family vs. non-family rms: A resource-based analysis of the effect of organizational culture.Entrepre- neurship Theory and Practice, 28(4), 363–381.

Zellweger, T., & Astrachan, J. (2008). On the emotional value of owning a rm.Family Business Review, 21(4), 347–363.

Zellweger, T., Eddleston, K., & Kellermanns, F. W. (2010).

Exploring the concept of familiness: Introducing family rm identity.Journal of Family Business Strategy, 1(1), 1–10.

Zellweger, T., Kellermanns, F. W., Chrisman, J., & Chua, J.

(2012). Family control and family rm valuation by family CEOs: The importance of intentions for transgenerational control.Organization Science, 23(3), 851–868.

Zellweger, T., & Sieger, P. (2012). Entrepreneurial orientation in long-lived family rms.Small Business Economics, 38(1), 67–84. The moderating role of socioemotional wealth55 123 R epro duce d w ith p erm is sio n o f th e c o pyrig ht o w ner. F urth er r e pro ductio n p ro hib ite d w ith out p erm is sio n.