etap_447 1051..1076 Governance, Social Identity, and Entrepreneurial Orientation in Closely Held Public Companies Danny Miller Isabelle Le Breton-Miller Based on notions from the identity theory, this study argues that in public firms in which ownership is concentrated, owner-chief executive officer (CEO) identities will influence entrepreneurial orientation (EO), and EO will relate to superior performance. Specifically, lone founder owners and CEOs will embrace entrepreneurial identities: their firms will exhibit high levels of EO and outperform. Post-founder family owners and CEOs, given their ties to family in the organization, will assume identities as family nurturers, thereby limiting EO and performance. Family firm founders will exhibit blended identities and demonstrate intermediate levels of EO and performance. This analysis of Fortune 1000 firms finds support for these arguments. Entrepreneurial orientation (EO) is a much-explored dimension of strategy making that has been found to arise in specific contexts and to have significant implications for firm performance (Green, Covin, & Slevin, 2008 Wiklund, 1999 Wiklund & Shepherd, 2005). Although there have been different conceptions of its components (Lumpkin & Dess, 1996), research has converged on three core dimensions of EO���innovation, risk taking, and proactiveness (Covin & Slevin, 1989 Miller, 1983). Each of these dimensions is said to contribute to the degree to which an organization is ���entrepreneurial��� according to earlier classical definitions of that concept (Collins & Moore, 1970 Knight, 1921 Schumpeter, 1934). Certainly, there has been ample research on EO over the past two decades (Cherchem & Fayolle, 2008 Rauch, Wiklund, Frese, & Lumpkin, 2009). Environmental and struc- tural correlates of EO have been investigated, revealing, for example, that EO tends to be more pronounced in uncertain environments and in firms utilizing organic structures (Becherer & Maurer, 1997 Covin & Slevin, 1989 Dess, Lumpkin, & Covin, 1997 Lumpkin & Dess, 2001). EO has also been shown to be related to market-driven strategies (Covin, Green, & Slevin, 2006 Green et al., 2008) and self-confident chief-executive- officer (CEO) personalities (Begley & Boyd, 1987 Poon, Ainuddin, & Junit, 2006 Please send correspondence to: Danny Miller, tel.: 1-514-340-6501 e-mail: danny.miller@hec.ca, and to Isabelle Le Breton-Miller at isabelle.lebreton@hec.ca. P T E & 1042-2587 �� 2011 Baylor University 1051 September, 2011 DOI: 10.1111/j.1540-6520.2011.00447.x

Simsek, Heavey, & Veiga, 2010). Most studies, however, focus not on the sources but rather the performance consequences of EO. Although findings conflict, on average, they suggest that EO tends to improve performance (see the reviews by Wiklund, 1999, and Wiklund & Shepherd, 2005). For major, publicly traded companies, which have too rarely been a focus of EO studies, a principal potential source of EO has been neglected���namely governance conditions (Miller, 2011). We shall maintain that in many such firms, the identity of major owners and owner executives may shape EO and thereby firm performance. Indeed, the myriad studies of the relationships between ownership concentration and performance, many based on agency rationales, have been plagued by disagreements���with some revealing that major owners benefit firm performance and others finding just the opposite (Morck, Shleifer & Vishny, 1988 Shleifer & Vishny, 1997). These conflicts also charac- terize the literature on family firms, both empirically (see, for example, Anderson & Reeb, 2003 and Villalonga & Amit, 2006 versus Claessens, Djankov, Fan, & Lang, 2002 Holderness & Sheehan, 1988 Maury, 2006 P��rez-Gonz��lez, 2006 and Miller, Le Breton- Miller, Lester, & Cannella, 2007), and conceptually (compare Miller & Le Breton-Miller, 2005, and Miller, Lee, Chang, & Le Breton-Miller, 2009 to Schulze, Lubatkin, & Dino, 2003, and Schulze, Lubatkin, Dino, & Buchholtz, 2001). This article will maintain that the identities of owners are very relevant to resolving these debates. Specifically, the social context of ownership will shape owner identities and hence their entrepreneurial preferences and the conduct and performance of their firms. We shall first lay out our general model of how role and social identities may be shaped by the different social contexts of those who govern. Then we shall generate hypotheses on how these identities develop and shape EO and firm performance in lone founder, family post-founder, and family founder firms, respectively. Theory Development: Ownership, Identity Theory and Social Identity Theory The Structure of our Model Our thesis is that there is an important connection between the social contexts of different types of major owners and owner-managers and the role identities and social identities they assume���identities that in turn shape their priorities for their businesses, and hence the strategies and financial success of those businesses.1 Our focus will be on three types of owners and owner-managers: those who are lone founders and those who are family owners or managers the latter may be either founders or nonfounders���and that is a second important contrast that distinguishes different types of family businesses and their conduct (Chrisman, Chua, & Litz, 2004 Miller et al., 2007). Lone founder identities will be influenced both by the roles they perform as business builders and by their desire to model 1. Others too have argued that owner values might shape corporate behavior (Arr��gle, Hitt, Sirmon, & V��ry, 2007 Greenwood, Deephouse & Li, 2007 Miller, Le Breton-Miller, & Lester, 2010, 2011). Some have focused on the social context of family owners. Naldi, Nordqvist, Sjoberg, and Wiklund (2007) did so in examining risk, in a sample of small Swedish family firms. Short, Payne, Brigham, Lumpkin, and Broberg (2009) found that family firms underplayed the dimensions of proactiveness and risk in their annual reports. These authors, however, failed to distinguish between family firms and those of lone founders. Miller et al. (2011) found that distinction to be important in their study of strategic investments to conserve or grow a business. They did not, however, focus on the risk inherent in or the boldness of those investments���aspects central to EO (Covin & Slevin, 1989 Miller, 1983). 1052 ENTREPRENEURSHIP THEORY and PRACTICE

themselves after successful members of a founder peer group. Based on concepts from identity theory (IT) (Stryker, 1987) and social identity theory (SIT) (Tajfel, 1974), respec- tively, lone founders will be argued to adopt the roles and social identity of an ���entrepre- neur��� business builder. Thus their firms will exhibit elevated levels of an EO and outperform. By contrast, non-founder family owners and executives of family firms will be influenced most strongly by their role-based personal relationships with other family members in their firm (Kowalewski, Talavera, & Stetsyuk, 2009 Schulze et al., 2003 Sciascia & Mazzola, 2008). IT suggests that given this familial role context in the business, these parties will adopt identities as ���family nurturers��� (Bertrand & Schoar, 2006 G��mez- Mej��a, N����ez-Nickel, & Gutierrez, 2001 G��mez-Mej��a, Tak��cs Haynes, N����ez-Nickel, Jacobson, & Moyano-Fuentes, 2007).That identity, we shall argue, will foster conservatism and resource extraction from the firm to serve the family, thereby limiting EO and performance. A third group of principals we shall distinguish is that of the founder of a family business, whose social context blends that of the lone founder and the family owner-manager, and hence whose identity and conduct will also be blended (Shepherd & Haynie, 2009). For those parties and the firms they own and run, our model suggests an ���average��� level of EO and performance. Figure 1 summarizes our model. We shall develop rationales for just how social identities manifest in the context of the lone founder, family post-founder, and family-founder firm���before exploring their stra- tegic manifestations. Identity Theory, Social Identity Theory, and Their Distinctions and Application Identity is a core concept that links social structure to individual behavior. Identifi- cation with particular groups and adoption of a related set of roles can shape loyalties, priorities, and conduct (Stryker, 1980 Tajfel & Turner, 1979). IT and SIT are related perspectives on the way in which the socially constructed self mediates between social structure and individual behavior. Whereas IT is based in role identification and has a legacy in sociology, SIT is based on group identification and has its roots in social psychology (Hogg, 1993 Hogg, Terry, & White, 1995). Role identity is enacted as people fulfill role expectations in interacting with valued others and manipulating ���the Figure 1 Governance, Social Context, Identity, and EO 1053 September, 2011

environment to control the resources for which the role has responsibility��� (Stets & Burke, 2000, p. 226). Social identity is assumed when one considers oneself a member of a valued group and sees things from its perspective, with or without direct interaction with that group (Ashforth & Mael, 1989). Thus although both theories rely on self-assignment into a social category that drives behavior, largely due to the social context and activities of an individual, there are important distinctions between the two theories: IT is based on identification with a salient role a person plays it materializes via personal relationships and roles. By contrast, SIT is based on projective identification with a group more than a role���a demographic group, for example���which one considers oneself a member of, that one may not be in contact with, and that has a clear in-group/out-group boundary that allows some experienced distinctiveness (Hogg et al., 1995 Stets & Burke, 2000). IT, as we shall see, suggests some constraints to EO within post-founder family firms, whereas both IT and SIT provide reasons for higher levels of EO within lone founder firms. For family owners and executives, the family is present in the business as an integral part of the social context: thus the very salient role of father, mother, or son to other members of the firm cannot help but shape identity and behavior in the business, and hence business priorities. IT is of direct relevance here, whereas SIT is less so as it concerns less intimate, more aspiration-based, or ���demographic��� associations (Hogg et al., 1995 Stets & Burke, 2000).2 Lone founders are shaped by their roles as entrepre- neurs and by the parties with whom they interact in that capacity. However, especially in large, major companies, founders may also be influenced by an aspired or projected social identity with their much celebrated peers���other successful entrepreneurs, as opposed to the less exalted group of ���mere��� managers or administrators. Here we are squarely in the province of SIT. We shall discuss briefly the manifestations, modes of activation, and motivations behind the identifications postulated by IT and SIT, and also show how they may influence the entrepreneurial behavior of family and lone founder owners. Identity Theory and Family Firm Ownership Our core argument in this section will be that given their close associations with other family members, family owners and managers will adopt family nurturer identities, which may limit EO within their firms. According to IT, people come to know themselves by interacting with others (Cooley, 1902 Mead, 1934 Stets & Burke, 2000). Thus the self is a product of social construction, not an autonomous psychological entity but rather ���a multifaceted social construct that emerges from people���s roles in society��� (Hogg et al., 1995, p. 256). Moreover, variations in self concepts are due to the different roles people occupy: mother, wife, and social worker (Stryker, 1987). Role identifications are self conceptions, self-referent cognitions, or self-definitions that people apply to themselves because of the role positions they occupy, and/or ���through a process of labeling or self definition as a member of a social category��� (Hogg et al. 1995, p. 256). But how do role identities become activated, what motivates their adoption, and how they are manifested? 2. Although SIT suggests that many family members will distinguish a family in-group from a complemen- tary out-group, the latter is rather a vague and heterogeneous entity. Moreover it is the roles that family members play in a family context that have the clearest behavioral implications, and those relate more directly to IT than SIT. 1054 ENTREPRENEURSHIP THEORY and PRACTICE

Activation. Commitment to an identity has been shown to be linked to the number of persons to whom one is tied by holding that identity, as well as the strength and depth of those ties (Stryker & Serpe, 1982). Where an individual is in contact with those to whom he is close, he is more apt to enact the role that is expected of him by those individuals. For example, given the importance of the parental relationship, a father will tend to behave consistently in a fatherly role with his children (Homans, 1950). Where these relationships are close and involve multiple family members, the identity associated with that role will be reinforced. Motivation. Identity theorists believe that a sense of self-efficacy comes to an individual as a result of properly performing a role���that is, fulfilling the expectations of valued partners. For example, a father may feel proudest and most satisfied when behaving according to the reasonable expectations of his wife and children. That sense of self- efficacy is a pivotal motivating factor contributing to role enactment and role identity (Burke, 1991 Stryker, 1968, 1987). Manifestation. Role-based identities manifest in differentiated and complementary rela- tionships with other members of a group. People recognize each other as role occupants and incorporate into their identities the ���meanings and expectations associated with [their] role and its performance��� (Stets & Burke, 2000, p. 225). Thus, for example, in a firm owned by several members of the same family, the roles they perform in the firm and just how they perform those roles will be influenced by family expectations and the roles they play in the family���as we argue below (Gersick, Davis, Hampton, & Lans- berg, 1997). Family Firm Ownership, Family Nurturer Identity, and EO. Major owners and owner- executives from the same family working together in their business tend to interact quite closely (Gersick et al., 1997 Miller & Le Breton-Miller, 2005). Their kinship lends to their associations the intensity characteristic of a primary social group: long duration, emotional closeness, and mutual dependence (Homans, 1950). Thus family members often will have a disproportionate impact on one another when acting within a business context, and may share a common familial identity, reinforced by the roles they play (Cruz, G��mez-Mej��a & Becerra, 2010 G��mez-Mej��a et al., 2007). Frequently therefore they will be influenced by family loyalties and bonds of familial reciprocity and altruism (Schulze et al., 2001). Family owners and executives will try often to fulfill the expecta- tions of family members close to them and will receive socioemotional satisfaction from doing so. Thus they enact their identities by embracing roles as family protectors and nurturers as well as company principals (G��mez-Mej��a et al.). Their nurturer/protector role identities may induce family members to see the business as a source of family financial security and reward, family reputation, and careers for current and subsequent generations (Gersick et al., 1997 Lansberg, 1999 Minichilli, Corbetta, & MacMillan, 2010 Schulze et al., 2003 Ward, 2004). Thus family members acting as major owners and/or executives of their business may, in their decisions, enact their roles as family protectors and nurturers by serving the family through the business, thus sometimes trading off benefits to public shareholders against benefits to the family (Kowalewski et al., 2009 Sciascia & Mazzola, 2008 Sraer & Thesmar, 2007). For example, they may favor stable, generous dividend payments, and perquisites for family members over reinvestment in the firm or rapid growth (Bertrand & Schoar, 2006). Resources might therefore be lacking for innovation, and risk taking would be avoided as it could jeopardize a steady stream of dividend income (Bloom & Van Reenen, 2007). 1055 September, 2011