Explaining leadership in family firms: Reflexivity,social conditioning and institutional complexity

Introduction

Research on leadership in family firms focuses on the ‘reciprocal influence of family and business’ (Zahra and Sharma, 2004: 333), viewed in the context of family and business logics (Friedland and Alford, 1991), when family and non-family CEOs exercise their interests within different operational and governance structures (Miller et al., 2011, 2013, 2014).

In an effort to explain the performance implications of family firm ownership (and leadership), such work has been guided by ideas taken from agency theory (Fama and Jensen, 1983; Schulze et al., 2003) and stewardship theory (Gomez-Mejia et al., 2011). To date, these theoretical frames have been ‘double-edged’ because they have been used to emphasize opposing facets of family leadership. In the case of agency theory, some scholars emphasize the reduced agency costs of aligned interests in family firms that bring commercial benefits while other scholars indicate how family CEOs ‘use their superior positions and knowledge to exploit less influential owners and to benefit themselves at the expense of the company’ (Miller et al., 2013: 554).

Efforts to reconcile these opposing insights – using agency theory – has led to work that focuses on the contextual conditions under which these parts of agency theory (and stewardship theory) pertain. So far, this work has indicated how a dispersed governance structure abates family-orientated distractions because it is assumed non-family CEOs exercise market discipline that has benefits if they work alone but are monitored by multiple major owners (Miller et al., 2014). In other research, scholars consider the combined effects of not just governance contexts but also firm size (administrative complexity). These studies indicate that attention should be given to different types of family firms as this shapes the extent to which family CEOs remove agency problems as opposed to precipitating family firm failure, as a consequence of poor management decisions (Miller et al., 2013).

Taken together, these research studies represent attempts to develop context-based approaches of leadership in family firms. Yet, while such work has contributed to understanding the performance implications of different family firm contexts there are limitations in the way this work is theorized. First, the distinction made between family and non-family CEOs (Miller et al., 2014) has been used as a vehicle to make distinctions between actors that adhere to either family or business logics; that is, non-family CEOs mitigate the non-rational decision making by family CEOs because they do not have allegiances to the family. We argue that this binary interpretation of actor interest over-simplifies ‘leadership’ as the actors are relegated to the mere ‘carriers’ of family and business logics.

Second, while ownership and administrative contexts shape decision making, the current formulation over-emphasizes the structural part of family firm governance confining agency to a second order concept. The explanatory utility of agency is sub-summed within the structural conditions because the configuration of interests (and action) is consigned to the relative mix of family and non-family membership (Miller et al., 2014) or it is attributed to the combinational effect of administrative complexity and ownership structure (Miller et al., 2013). In this formulation, the actions of actors are aligned with structural contexts rather than reflecting a negotiated process through which circumstances are weighed-up by individuals.

Lastly, the idea of social complexity is only partially revealed in the study of business and family logics. Although logics have been used to define social context in family firm research (Greenwood et al., 2010; Miller et al., 2011) these ideas have not been used to their full explanatory potential. This is because the focus on the performance-context link has led to a very narrow interpretation of social complexity that ignores other societal orders (Friedland and Alford, 1991; Thornton et al., 2012). The limitations are magnified when agency is relegated to a second order concept because the focus on firm performance black-boxes action within a narrow interpretation of social complexity and family firm operations.

In this theory article, we reassess the relationship between agency, organization and society in the study of leadership in family firms. To begin with, at the highest level of abstraction are the institutional logics shaping the cognitions and actions of actors, of which there are many and not just family and market (Friedland and Alford, 1991; Thornton et al., 2012). Action, including leader agency (and the acts of followers), cannot be explained without reference to these social orders, but nor do these determine action. Leadership in family firms occurs as actors operating through organizational structures reflexively assess the complexity around them. This is not a simple top-down set of relationships, rather this relationship is nested because although ‘logics have the power to impede or facilitate the action of different actors the activation of such powers is contingent upon those agents who conceive of and pursue what might be possible’ (Delbridge and Edwards, 2013: 931). Reflexivity counts in leadership because action is not just an outcome of the social conditioning of logics or the actor positions of individuals in firms; it is also a feature of how actors come to understand their place in the world (Archer, 2003, 2012; Delbridge and Edwards, 2013).

In contrast to the existing work on leadership in family firms that struggles to explain the strategic choices made by family CEOs, our revised approach focuses on their ‘institutional biography’ (Suddaby et al., 2012) rather than basic distinctions made between family and non-family heritage (Miller et al., 2014). Actor biographies are not confined to the influence of family or market logics but reveal cognitive registers related to other social orders such as, for example, the professions (Abbott, 1988) and communities (Campbell, 2011). How these come to shape action depends on not just the operational and governance structures of the firm (although these are necessary considerations); strategic choices also rely on the individuals who make ‘normative judgments among alternative trajectories of action, in response to the emerging demands, dilemmas, and ambiguities of presently evolving situations’ (Emirbayer and Mische, 1998: 971). This is when actors assess their actions, which is not pre-programmed but confirms reflexivity ‘as the process mediating the effects of our circumstances upon our actions’ (Archer, 2012: 6). This explains why it is that some family firm CEOs see events as opportunities for socioemotional wealth generation while for others the same moment appears to present significant obstacles to financial performance. This also explains why ownership and operational complexities are poor predictors of leadership behaviours when the institutional biography and reflexivity of the actor are ignored.

In this article, our first contribution is to consider ‘leadership’ in family firms as a relational feature of the linkages between reflexive actors, organizational contexts and multiple social orders (Delbridge and Edwards, 2007, 2013; Mutch et al., 2006). This theorization relies on an institutional logics framing that draws attention to the nesting of leadership within organizational and social contexts when each level of analysis is treated as a distinct yet integrated feature of the social conditioning of strategic actions. This framing offers an opportunity to make a further contribution, which is to debates about the ‘lopsidedness’ of family firm studies that have emphasized the market at the expense of the family (Jennings et al., 2013: 25). We reflect on the relationship between the reflexive individual and the scope for leadership action that extends beyond the family-business dynamic.

Our final contribution concerns the notion of complexity in leadership studies and how this is theorized. Recent developments in the area of complexity leadership theory indicate that in the post-industrial era social contexts reveal adaptive systems that constitute ‘an emergent, interactive dynamic – a complex interplay from which a collective impetus for action and change emerges when heterogeneous agents interact in ways that produce new patterns of behaviour or new modes of operating’ (Uhl-Bien et al., 2007: 299; emphasis in original). Advocates reject the idea that leadership refers to the actions of a ‘leader’; instead, the explanatory focus shifts toward complex interactional dynamics from which adaptive outcomes emerge.

We want to caution against the view that leadership is contextually contingent and culturally relative such that these interactional dynamics are a necessary condition of post-industrial society (Du Gay, 2003; Edwards, 2011; Sturdy and Grey, 2003). In the context of family firms it is vital to understand how actors reflexively engage with social complexity; these are not self-evident certainties about post-industrial society but reveal reflexivity as the process through which institutional processes influence the subjectivity of actors. We assess these ideas in relation to the reflexive imperative in late modernity as it relates to leadership in family firms (Archer, 2012).

The article is structured as follows. We begin by considering recent work that helps to outline the idiosyncrasies of the family firm, leadership and strategic action. This gives the impetus in the next section to reassess the linkages between leadership and social complexity; that is, we follow our critique with a theoretical assessment of the relationship between actors, organizations and society. We use this to outline a nested view of leadership and how this relates to family firms by indicating the way actors cope with social complexity. We develop these ideas further in the last two sections by, first, considering our framing in relation to the family firm literature and, second, using these ideas to critique work on complexity leadership theory to qualify how complexity and reflexivity relates to family firms. The final section concludes.

Leadership and family firm research

In a recent research study of leadership in family firms, the authors state that despite ‘numerous empirical studies doubts remain as to whether family ownership and family management are good or bad for a business’ (Miller et al., 2013: 553). This statement reveals two trends that have dominated recent discussions about leadership in family firms. First, there is growing interest in examining the contexts within which the idiosyncrasies of family firm ownership and leadership positively impact financial performance and, second, this work reflects interest in framing leadership in terms of the market rather than the enterprising family (Jennings et al., 2013). The upshot is that leadership studies in family firms have largely developed in ways that prioritizes instrumentalism when social complexity is assessed in terms that focus on the linkages between organizational forms and financial performance (Miller et al., 2011, 2013, 2014).

Although such research recognizes the importance of family orientations, these attributes are interpreted in ways that emphasize how leaders might better use their familial assets for economic ends. With long tenure and professional knowledge the founder exerts a huge influence on the culture and performance of the firm (Kelly et al., 2000; Ogbonna and Harris, 2001; Schein, 1985). The close relationships and repeated interactions among family members are associated with in-depth knowledge about their products, customers and competitors, whose transfer and renewal across generations represent a key asset (Cabrera Suarez et al., 2001). The problem, as it is presently stated, is to know when such assets might lead to long-term investments or when they will feed a preference for non-economic interests, such as ego building (Gomez-Mejia et al., 2007) or serving personal (Schulze et al., 2003) and familial interests (Betrand and Schoar, 2006; Bloom and Van Reenen, 2007).

Attempts to explain when the benefits of family ties coincide with improved firm performance rest on contextualized studies. For example, Miller and colleagues (2013) argue that high performance family firms are those that exhibit a modest level of administrative complexity coupled with highly integrated ownership structures that involve only a few family members. Firm size is used as a proxy to explore the worth of the tacit knowledge of the family CEO, compared with the professional management of non-family CEOs, and a concentrated firm ownership signifies family contexts that reduce the social burden put upon CEOs, mitigating family conflict. Leadership in family firms is assessed according to operational and governance complexities, but in doing so this work effectively reifies those social contexts; that is, they impose a top-down interpretation of the context that is used to explain when the negative impact of family values might be mitigated.

We would question whether firm size is a reasonable proxy for the knowledge benefits attributed to family firm ownership or whether a concentrated ownership structure mitigates conflict. The problem is that, in trying to account for the linkages between family firm characteristics and performance, scholars conflate the ownership structures of the firm and action not giving due attention to the motivations of those involved or how decisions follow a trajectory particular to the firm in question. Complexity is, as a result, confined to a snapshot of the firm, defined in narrow terms, which effectively ignores the agent who is making the strategic decisions. The agent is relegated within this formulation to a simple carrier of specific logics because the conditions – modest administrative complexity and close family relations – are prioritized over the active deliberations and actions of the actor.

We would argue that the strategic motivations of family CEOs are likely to be more complex than is thus far outlined. This is because, in an effort to make use of mid-range theories such as agency theory to explain firm performance, scholars have tended to look inside the firm excluding social complexity more broadly defined and the executives themselves. This is a problem because family leadership is shackled to a very narrow interpretation of action, orientated as it is to market performance, which ignores other cognitive registers. To overcome this market-bias and therefore ‘open-up’ explanation of family leadership, it is necessary to explain social complexity as a nested set of relationships.

Social complexity and family firms: An institutional logics perspective

The motivation to overcome the shortcomings of existing leadership studies in the family firm literature is driven by a desire to develop a stratified approach, which sets out to explain the relationship between complex social orders, organizations and human action. Based on Friedland and Alford’s (1991) work, we explore leadership in the context of ‘individuals competing and negotiating, organizations in conflict and coordination, and institutions in contradiction and interdependence’. This is a nested view of social conditioning and complexity because organizations and logics ‘specify progressively higher levels of constraint and opportunity for action’ (Friedland and Alford, 1991: 240–241, 242) with each institutional order equipping ‘individuals with vocabularies of motives and with a sense of self’ that frames ‘not only that which is valued, but the rules by which it is calibrated and distributed’ (Friedland and Alford, 1991: 248).

According to Thornton et al. (2012) there are multiple societal orders including the state, market, family, religions, professions, corporations and community that are ‘organized around unobservable substances where constellations of particular practices are understood as their enactment’ (Friedland et al., 2014: 3). However, logics cannot be reduced to these practices (as pre-existing categories) because they only materialize into expressions of social relations at the point when they obtain as objects, practices and subjects. Logics rely on the subjective pursuit of a value and one only gains access or becomes emotionally attached when people build materialized languages around these objects and perform them through action.

Correcting substantialism that reifies social artifacts, this theorization of logics and social conditioning relies on a metaphysical category: an institutional substance:

[L]ogics depend on making the invisible substance visible. Institutional practices are the visible face and the condition of possibility of institutional substances, and hence the source of their identity across time, but reciprocally, the continuity of those practices depends, particularly at moments of logical multiplicity, instituting and institutional failure on a metaphysical belief in the substance, as an actionable good. (Friedland et al., 2014: 2)

In the context of leadership in family firms, this formulation provides the basis to correct recent assessments of market and family logics in family firms (Miller et al., 2011). To begin with, it is necessary to recognize social complexity is not just limited to family and market logics and this means that leadership decisions are informed by values that extend beyond the boundaries of the family and administrative structures of the firm. The evidence implies that family firms are often tightly integrated within, for example, local communities, when such relations are based on ‘unity of will, belief in trust and reciprocity’ (Thornton et al., 2012). Mutually supporting relationships among employees, suppliers and customers (Johannisson, 2011; Miller et al., 2009) can often bring benefits to family owners, including information and resources on favourable terms and personal prestige (Campbell, 2011). Indeed, the co-existence of multiple logics shape business practices in ways that expose how non-market values inform commercial decisions, which is seemingly apparent in the way ‘many communities have mobilized resources and energy to prevent mass-market firms such as Wal-Mart and or Starbucks from opening stores that might threaten long-standing local [and family] establishments’ (Marquis and Lounsbury, 2007: 800).

Our interest in explaining the connections between multiple social orders, the organizational context and agency is further demonstrated with reference to low-value added (LVA) family firms. In LVA firms, recruitment is influenced by poor access to the labour market, which means that the search for employees is based on ‘friendship or family ties rather than instrumental rationality’ that emphasizes the importance of informal ties rather than hierarchical structures (Edwards et al., 2006: 710). LVA firms are also more likely to conduct their business in localized markets so they influence prices and it is because they serve a community that actions reflect integration. The combined role of context and organizational forms can be expressed using an ideal type – the fraternal firm:

[T]he [fraternal firm] has an established niche (for example in the building trade), so that it has some control in the product market. Workers are hired through (restricted) word-of-mouth means, family resources help to sustain the business, and workers are treated as the equals of managers (so that the style is participative) . . . The firm is the property of the owner-manager, and workers who question the brotherly and harmonious image may be forced out. ‘Participation’ means that workers’ preferences are treated seriously and that workers are not treated as mere factors of production. Rules, moreover, reflect the idiosyncrasies of individual owners and are likely to be [shaped by personal and kinship relations] rather than formalized. (Edwards et al., 2006: 712; emphasis added)

The actions of LVA firms confirm higher orders based on family ties, because family loyalty is an orientating principle, but so is ethnicity and community because the business transactions of actors, including recruitment, is a feature of the strong ties and unity within this context. This means ‘community’ holds a specific value and this ‘multiplicity’ more closely resembles complexity because this sits alongside market and family considerations (Friedland et al., 2014). The operational structures of LVA’s and broader social relations shape how this complexity is understood, but this does not determine action because owner managers navigate complexity. Leadership is not a simple market or family calculation, mediated by specific firm structures but involves an active engagement in value judgements that cannot be reduced to the logics or family firm structures. Social complexity is apparent in how different logics are co-implicated while leadership has to be explained in relation to progressively higher levels of opportunity and constraint, which depends on the deliberations of actors.

The problem with leadership studies in family firms is that little attention is given to explaining the ‘idiosyncrasies of individual owners’ (Edwards et al., 2006: 712). This is necessary because social complexity inculcates people within a highly deliberative process; how they engage in action in part depends on how they weigh-up complexity and make strategic choices. Take the example of entrepreneurial bricolage (Baker and Nelson, 2005). This work indicates how entrepreneurs, when confronted with resource constraints, recombine resources in ways that are novel and otherwise uniquely useful. Bricolage involves a process of making do when entrepreneurs apply ‘combinations of resources at hand to new problems and opportunities’ (Baker and Nelson, 2005: 333). This is significant because in making do the entrepreneurs break prevailing definitions of their physical and institutional world. They make choices about their actions in relation to the value they attribute to such work, which confirms their ability to resist the constraints of their local environment, especially the market (Phillips and Tracey, 2007).

A feature of this work has been to outline the context within which bricolage occurs and whether this occurs as a selective or parallel process. The entrepreneurs that were involved in bricolage did a considerable amount of scavenging, making do with whatever tools, materials and items they could lay their hands on. In the case of actors involved in parallel bricolage, what is interesting is that these entrepreneurs were less likely to have been formally educated or trained. Instead, they developed ways of working that reflected ad hoc solutions for combining and extracting new services from their resource troves. These entrepreneurs consistently tested and/or challenged codes of operating and professional norms of practice by relying on the cooperation and willingness of their customers to let them get away with it. They overcame limitations in resources by being innovative (they could cobble solutions together) and because they had close, reciprocal relationships with members of their community that presented opportunities to ignore professional norms.

For those involved in selective bricolage, they were not locked into the local communities because the markets were more competitive. The entrepreneurs reverted to conventional business practices and professional principles because the value they attributed to market principles were realizable when local market conditions changed (Baker and Nelson, 2005). Distinctions between parallel and selective bricolage infer alternate interpretations of the value of ‘making do’. But these distinctions cannot be explained by the social orders or firm operations alone because decisions to follow a parallel or selective approach depends on the entrepreneurs who make decisions about what matters to them. This view of agency rests on a specific ontological framing of social action that recognizes agency as a separate, albeit integrated, level of analysis. This counters agency theory as a mid-range concept because agency is conflated with organizational structures. Rather, to explain how actors navigate social complexity, we must explain why it is that some make strategic decisions that confound market values or emphasize community ideals over and above professional norms.

Agentic orientation and reflexivity: Leadership and social conditioning in family firms

The argument that has been presented so far has been to propose a stratified view of social action that provides the means to overcome the limitations of existing family firm studies of leadership. This has meant mapping out an inter-institutional system comprising of multiple logics that shape the cognitions and actions of actors. These societal orders are not felt directly by actors but are filtered by organizational forms and local contexts that present actors with chances and constraints for action. Our approach recognizes that within such circumstances actors make decisions about how to proceed. Strategic choices do not reflect logics directly but reveal a negotiated process that is only partially explained by the family firm context. The contingency of family leadership studies is replaced by a view that recognizes actors have interests and these only become meaningful as object-practice-subject materialize (Friedland et al., 2014).

Now that the full extent of social complexity has been made apparent, we still have to explain how actors navigate complexity. This means explaining (i) how actors connect with such complex settings and (ii) how they evaluate the opportunities for action given these contexts (Delbridge and Edwards, 2013). To understand how actors cope with complexity, we explore the modes of embeddedness or opportunities for action (Emirbayer and Mische, 1998). The modes infer three agentic orientations that constitute an iterational, projective and practical-evaluative mode. An iterational mode infers the habitual constitution of society that sustains established arrangements over time. A projective mode confirms the potential for actors to think imaginatively and conceive of alternate configurations, while a practical-evaluative mode indicates an ability to overcome ambiguities emerging in day-to-day events choosing between different trajectories of action (Emirbayer and Mische, 1998).

This line of analysis allows us to theorize the opportunities and constraints for action given the different ways actors connect with multiple logics. Logics obtain materiality through the actions of actors in organizational and wider social settings as circumstances unfold. This is evident in the case of selective bricolage when local contextual realities shift (market opportunities emerge), signaling a move (or the potential to do so) in the orientation of entrepreneurs depending on whether they see events as a chance to think of new possibilities (projective) or might require a form of social repair (practical-evaluative). Whether social circumstances reflect iterational, practical-evaluative or projective modes is not determined by social conditions but relies on the actors making a judgement within a meaningful context. Actors make decisions and to do this they are required to reflexively monitor their relationship with the social orders that shape their world. In this sense, leadership in family firms is a process of agentic collaboration that is ‘communication-centered’ (see Fairhurst and Connaughton, 2014) and leads to particular pragmatic outcomes (Raelin, 2011).

Reflexivity signifies that ‘agents have to diagnose their situations, they have to identify their own interests and they must design projects they deem appropriate to attaining their ends’ (Archer, 2003: 9). This undermines the explanatory utility of the existing contingency-based studies on family leadership because reflexivity represents a necessary additional process in the relationship between firm contexts and action. In particular, reflexivity is mediated by the way actors are ‘orientated’ toward the past, present and future, and, because this is not an innate human capacity, reflexivity also reflects the past experiences of actors (Suddaby et al., 2012).

To explore these ideas, we draw on the work by Archer (2003) who specifies several forms of reflexivity. Communicative reflexives are those who assess their place in the world inter-subjectively. For Archer (2003), these actors see themselves in relation to those around them and, in particular, those they see as crucial to maintaining the status quo. In the case of family firms, such embeddedness is not just restricted to family members but may also reflect community relations (Fairhurst and Connaughton, 2014). This feeling of closeness informs how individuals interpret what matters; for example, ensuring when family and community values trump market and professional rules (Baker and Nelson, 2005). Likewise, the involvement of non-family CEOs who are monitored by family members suggests that inter-subjectivity might also result in performance benefits if those involved recognize the reciprocal benefits of family and market knowledge (Miller et al., 2014).

In contrast to communicative reflexives, Archer (2003) outlines the particulars of what she calls autonomous reflexives to reveal a different reflexive person. These individuals think about what matters at a distance from existing arrangements. Rather than rely on individuals around them to confirm what should be done and how this might be achieved, such reflexives operate independently having established their own personal projects away from the existing arrangements. Such projects are different from those of communicative reflexives because the reflexive moment confirms a different connection with the past based on ‘contextual discontinuity’. This means the personal biography of the individual shows the influence of multiple discordant social orders. Mutch’s (2007) study of the Victorian entrepreneur Andrew Barclay Walker draws attention to Walker’s upbringing and how his often-disrupted childhood, family ties and Presbyterian education provided him with the foundations to envision new ways to manage public houses in Liverpool. The ability to see these opportunities and to act upon them say something about the connection between Walker’s ‘institutional experiences’ and the context around which public houses were run in Liverpool in the 19th century. Such people are the institutional entrepreneurs of society who find ways of doing things contra expected business practice.

The idea of self-monitoring is developed in Archer’s (2003) category of meta-reflexives. These are individuals that not only monitor their personal projects but also review that process. Similar to autonomous reflexives, these actors have experienced contextual discontinuity in their lives; but these individuals are also ready to remove themselves from situations where they feel unable to achieve their desired ends. This is relevant for understanding leadership in family firms because owner managers can make decisions that confound social arrangements. The study of the third generation business owner Thomas Berger (Dawson and Hjorth, 2012) is insightful because his decision to prevent the transfer of the business to his son and instead establish a trust follows concerns his son would not uphold the family values of the firm. Reflexivity is important because it has the potential to explain those times when embedded values such as family loyalty are challenged. We can see this in selective bricolage because the choice of actors to withdraw from ‘making do’ serves to highlight how the self-monitoring process ensured the interests of these actors were better served.

Attention to different forms of reflexivity presents an opportunity for family firm scholars to explore differences between and within family firms in relation to past experiences and the historical development of a firm. Pinpointing the emergent features of action is needed because current explanations of family firm behaviours have failed to shed light on ‘how family systems affect opportunity emergence and recognition’ (Aldrich and Cliff, 2003: 593). Combining agentic modes and reflexive categories has the potential to explain leadership in terms of the relationship between social contexts (revealed through different agentic modes) constituting multiple logics and the actors who assess the social context based on their reflexive capacities. These ideas help in the development of family leadership research because explanations that are based on ownership and administrative complexity ignore actors or how context-reflexivity relations shape action. We reject the idea that leaders carry logics because, while family and non-family executives are likely to have different interpretations of situations, it is wrong to assume that family or non-family heritage determines action. Recognizing different agentic modes allows us to theorize leadership as conditioned action (not determined) that plays out in relation to the reflexive abilities of the actors (family and/or non-family) who collide with the multiple social orders that constitute the social context.

Applying these ideas offers an opportunity to explain family leadership as a relational feature of the agentic mode and reflexive abilities of those involved in such action. Where actors are positioned iterationally (they are deeply embedded) and they are highly socialized within this setting (communicative reflexive) then leadership choices are likely to be framed by concerns to maintain the social and organizational status quo. Where there is a dominant logic (family) this is relatively straightforward because the combination of this orientation and reflexivity will be self-sustaining. Leadership choices will be directed toward the maintenance of the family logic. Where the social context shifts it might be expected that the resulting practical-evaluations are directed toward repairing that social context so the family logic is maintained. Changes in the context need not lead to new strategic directions because such decisions are considered in relation to the reflexive outlook of owners. On occasions, when decisions are conducted inter-subjectively between the family owner and those family members around them, then this is likely to re-enforce existing preferences. We can see this in the levels of inertia that affect decisions to divest in family-owned firms; the ‘combination of past successes, emotional attachments and path dependencies’ confirm the reluctance of owners to relinquish certain values and material resources in relation to the family structures embedded in the firm (Sharma and Manikutty, 2005: 293).

The strategic judgements of leaders will be far less obvious when the social context is defined by institutional complexity. We feel this is where our conception of conditioned action will present new opportunities for family firm scholars interested in leadership and strategic action. When multiple logics are at play – this is endemic and not the exception – then the ‘leader’ has to make more active, reflexive choices in relation to their context. Such reflexive moments are likely to consist of practical-evaluative or even projective modes because of competing logics. On such occasions, the strategic judgements of individuals will depend on the relative reading of social situations, which will be reconfigured in relation to past experiences.

We would suggest, for example, that those entrepreneurs engaging in parallel bricolage appear to resemble Archer’s communicative reflexives because they were integrated in the community and therefore enabled to act far more strategically. This is not the same for those engaging in selective bricolage because they reviewed this process changing their strategy as commercial opportunities arose. The ability to cut social bonds is both a feature of institutional complexity and the reflexivity of those involved, and, in this case, such moments were conducted at a distance from those around them. Rather than selective bricolage being conducted inter-subjectively (locked-in), this shows the self-monitoring of meta-reflexives who reverted to commercially-based operations.

As these illustrations infer, there are many possible combinations of context-reflexive relationship, which we feel has implications for addressing the bias toward market interpretations of family firm behaviours (Jennings et al., 2013) that includes the recent work on leadership in family firms (Miller et al., 2011, 2013, 2014). In the next section, we evaluate this reformulation to consider how this framing: ‘social contexts – personal preferences’ (Archer, 2012: 7) enables us to reevaluate leadership in family firms and how this pertains to social complexity.

Reflexivity, leadership and contributions to the family firm literature

Family firm scholars have reacted to the imbalance (toward the market) in studies by calling for ‘the resurrection of theories from family science’ with the aim of ‘enriching current understanding of how family affects business’ as well as ‘stimulating lesser-asked questions about how business affects family’ (Jennings et al., 2013: 25). It is in this context we see space for work on reflexivity and leadership because these ideas broaden the horizon of family scholarship and in ways that frame leadership in the context of multiple social orders. In particular, we offer refinements in the area of systems theory as it has been applied in family firm (Whitechurch and Constantine, 2004) and leadership studies (Uhl-Bien et al., 2007) to re-calibrate recent interpretations of social conditioning and leadership.

In the family firm literature, the adoption of systems theory has assumed the integration of individuals within the whole family system. This means, ‘the behaviour of individual members, when looked at separately, will not provide an accurate picture of what is occurring’ (Jennings et al., 2013: 27). Context is important and the family unit defines this. Family firm scholars using a systems approach also assume individuals are capable of self-reflection able to look at past decisions and consider alternate options for the future (Whitechurch and Constantine, 2004). In combination, systems theory confirms the need to consider social contexts and the reflexive actions of family members but there are drawbacks because the system is closed, confined to ‘the family’ while reflexivity is assumed to be a default condition that is equally and consistently exercised by all members of the family.

These limitations are only partially overcome in the family-business literature because, while the family system is ‘opened’ to include the market, this ignores other social orders. In turn, while self-reflection is broadened to reflect business concerns, the agency of actors is conflated with the structural conditions imposed by the family firm and so we lose sight of the actors engaged in action. Indeed, we have little idea about the breaks as well as the opportunities for reflexivity other than that leadership decisions are highly contextualized (Miller et al., 2013).

Our complaint is that unless we explain this dynamic beyond basic principles it becomes hard to explore the checks and balances that shapes strategic action. The preceding critique recognizes that leadership and our knowledge of the social world is open to continued revision. Social complexity exposes those multiple orders that exist in contradiction and interdependence and which are valued by individuals in different measure depending on the social context. Family firm forms specify the positioning of actors who engage in action because they constitute the ownership and governance contexts wherein actors form management preferences. But rather than assume that all actors are equally reflexive, our formulation indicates that leadership decisions reveal a sense of self that only emerges where the past meets the present and where different social orders collide, at which point actors decide on what matters. This may mean something very different as events unfold and depending on how they reflect on their position in society. Not all reflexive moments are conducted inter-subjectively and, as such, strategic decisions may run counter to established social arrangements, which are reflected within existing family firm structures.

What we are proposing is a position- and action-based approach, which offers a refinement of existing systems theory approaches in the family firm literature. The adoption of ‘social contexts – personal preferences’ allows us to reframe questions about leadership and complexity in family firms. We reject simple binary assessments of family leadership because with social complexity come the seeds for contextual discontinuity and incongruity and this can make communicative reflexivity that much harder to accomplish when the idea of the family is disrupted. Leaders may be forced to reflexively monitor situations ‘at distance’ as they have to decide between different social orders, as in the case of selective bricolage. Then again, social complexity can amplify co-existing orders (i.e. as with community orders) and in ways that support values (Greenwood et al., 2010). Rather than limit assessments to the way family can enrich business, we suggest that scholars need to be sensitive to the way multiplicity, and not just family and market, shape deliberations conducted inter-subjectively. In the case of LVA family firms, the findings suggest the idiosyncrasies of the family owner cuts across and beyond family boundaries to include ethnicity or religion and also the community. By broadening the idea of ‘family system’ we open up space to look at leadership decisions in the context of multiple, co-terminus logics that amplify family values, creating contextual continuity in the midst of potential discontinuity.

We suspect this dynamic will inform debates about the differences between family firms (Melin and Nordqvist, 2007) and sustainability of the family (Stafford et al., 1999). Such work is likely to benefit from the current extension because different modes of reflexivity shape how unconditional loyalty is interpreted in complex situations. With multiplicity comes important issues about the way ‘family leaders’ reflect on what constitutes ‘family values’. This will help us to overcome the limitations of previous family leadership studies that use the family–non-family rubric as a proxy to predict leadership actions because reflexivity precedes the application of tacit knowledge; that is, reflexive actors variously decide how they want to use their knowledge resources (Archer, 2012). Strategic decisions are not determined by actors as ‘family’ or ‘commercial’ agents but as individuals confronted by complexity and informed by different personal preferences. This explanatory step will help overcome the lopsidedness of family firms studies because such an approach extends the scope for understanding leadership and it does so by assessing the motivations and emotions that lie behind different ‘social context – personal preference’ dynamics.

Reflexivity and complexity leadership studies in late modernity

Our interest in the relationship between social complexity and reflexivity also has implications for the way leadership might be theorized more generally and how this feeds back to our basic set of arguments in terms of a systems approach.

Tracing developments within the general leadership literature, it is possible to observe a move from studies that focus on leadership within a bureaucratic framework toward those that are securely anchored within a post-bureaucratic interpretation of economic society (Uhl-Bien et al., 2007). In the bureaucratic form, leadership studies embrace essentialism, examining leader behaviours, their actions and thoughts, and the influence this has on organizational outcomes (Bryman, 2004). This economic-actor model of leadership, which is often grounded in monologism (Cunliffe and Eriksen, 2011; Fairhurst, 2009), attempts to explain how leaders can best maximize employee efficiency, minimize work conflict and recognize opportunities for securing economic returns (Alvesson and Sveningson, 2003; Tourish et al., 2010). The omniscient leader (albeit working with other actors) defines what is important – profit maximization – and it is by their actions they deliver these ends by ensuring workers achieve psychological and social identification with the organization (Collinson, 2005).

Criticisms tend to take two forms: (i) normative-based accounts rarely deliver ‘explanatory constructs good for all situations’ (Bass, 1997: 130); and (ii) such studies ignore the challenge of explaining how social complexity impacts strategic action. An attempt to remedy the second criticism is apparent in complexity leadership theory, which builds on insights offered from the complexity sciences (Anderson, 1999; Brown and Eisenhardt, 1997; Goldstein et al., 2010; Marion and Uhl-Bien, 2001). This conception emphasizes networks of interaction consisting of ‘complex patterns of conflicting constraints, patterns of tension, interdependent relationships, rules of action, direct and indirect feedback loops, and rapidly changing environmental demands’ (Uhl-Bien et al., 2007: 307). As such, leadership is viewed as part and parcel of the interactional dynamic that exists among interdependent agents drawn together in the post-industrial milieu. The aim is to explain ‘how enabling leaders can interact with the administrative superstructure to both coordinate complex dynamics (i.e., adaptive leadership) and enhance the over-all flexibility of the organization’ (Uhl-Bien et al., 2007: 302; emphasis added).

The move from essentialist to interactional framings of leadership presents an important step in releasing leadership scholarship from rational actor models, which includes studies of family leadership. However, there is an inherent tension in this work because interactional framings have failed to release leadership from the normative readings found in the former. Boal and Schultz (2007) build on the idea that, within the context of networks of interaction, effective leaders can stimulate creativity by channeling knowledge. By promoting organizational narratives about an organizations history, such activities can cultivate emotional intelligence, including self-awareness, vulnerability, empathy and an openness to learning from others regardless of their positional authority. Key is the ability of those individuals to act in adaptive ways irrespective of the power dynamics of the social situation (Boyatzis and McKee, 2005; Goleman, 2013; Goleman et al., 2008).

The problem is that there is little in the way of conceptual architecture explaining how leadership becomes enabling. The belief that complex adaptive systems can be catalyzed for economic ends is developed at the expense of explaining how actors achieve this (Levy, 2000); that is, social complexity works behind the backs of people according to the universal rules of self-organization, emergence and connectivity. This problem is made worse in the way social context is treated. For complexity leadership scholars, the social context is an open system that confirms the dominance of the knowledge economy. Yet, this seems hard to countenance because the knowledge economy is taken to be a ‘given’, such that leadership is discussed in terms that emphasize what ‘ought’ to happen. Unfortunately, what ‘ought’ to happen rarely equates with what does happen. While elements of post-industrial society might shape organizational action the nature of such influences is neither wholesale nor even (Du Gay, 2003), which is what we see in the LVA family firms and entrepreneurial bricolage.

Our understanding of late modernity is different to that given by complexity leadership theorists and the difference hinges squarely on the social context – personal preference dynamic (Archer, 2012). In terms of context, for complexity leadership theorists, the inevitability of the post-industrial era has led to the ‘flattening’ of civil society, removing the structural constraints that shape human action. The impact of this transformation, what has been termed ‘reflexive modernization’ (Beck et al., 1994), has been to create a world where individuals are far more reflexive because past social structures have been dissolved, and this has allowed for the flow and recombination of ideas in ways that are unfettered in the new ‘open’ society.

The counter point, which is where we stand, is ‘the general intensification of reflexivity . . . is directly related to mutually reinforcing changes in cultural and social cultures’ (Archer, 2012: 3–4). Rather than reflexivity emerging as an imperative tied to a flatter society, which feeds a normative reading of the knowledge economy, we suggest that reflexivity has changed in so far as modernity is being characterized by more, not less, social complexity. Indeed, as the examples in the family firm literature suggest, reflexivity is situated and such localization draws into relief how past social structures inform present decisions creating space to reject existing professional and market rules (Baker and Nelson, 2005). The dynamics of the knowledge economy are not given; instead, reflexivity reveals the challenge of ‘managing’ complexity among co-existing structures. As such, we agree with Archer (2012: 1) when she argues, ‘for the first time in human history the imperative to be reflexive is becoming categorical for all’.

This is important for the study of family leadership because reflexivity is the mechanism through which we come to understand the impact of social complexity on enduring social structures, including the family. Reflexivity is the main driver because it is through human action that such processes of change take hold. We contend that social complexity is ‘structural’ and rather than the reflexive moment being opened-up, as argued by complexity leadership theorists, we argue that it represents different modes and challenges because of the way multiple social orders impact agency. This does not mean that family, community or market orders have ‘disappeared’, rather the idea of ‘family’ has to be assessed in light of multiplicity and how individuals make sense of it (the opportunities and barriers) should be the broad focus of future family leadership studies.

In developing ideas around different ‘social context – personal preference’ modes we caution against the ‘open’ society view assumed by complexity leadership theorists and, as such, we caution against framing family leadership in ways that ignore the deeply embedded nature of conditioned action. Rather than treat reflexivity as a seeming homogenous condition of late modernity, we propose that distinctions in the reflexive mode present opportunities to explain variation in the strategic actions of family leaders.

Conclusion

Within the family firm literature we see opportunities to extend existing work to investigate the reflexive capacities of family firm executives. Moves to open up the idea of reflexivity will allow scholars to comment on the origins of competing family firm behaviours, which continue to challenge scholarly investigation. Such work will have to be historical (Thornton et al., 2012) in so far as the reflexive moment has to be explored in the context of the biography of owner-managers (Mutch, 2007), and it will have to include orders beyond the family and the market to adequately reflect social complexity in the context of late modernity (Archer, 2012). The contingency studies of existing family leadership studies are rejected in favour of historical-biographical investigations of leadership that must explain the necessary connection between social context and personal preferences. Our aim has been to emphasize family leadership as action taken by reflexive actors within a meaningful context. In doing so, we want to emphasize that not all individuals are equally reflexive and this must be assessed as a relational condition of the situated location of the family firm.