Heroes of the Green Room – How able successors revitalize and reinvent family firms

ABSTRACT

Drawing on contingency and upper echelon theories, this study conceptualizes family firm CEO succession as a special window of opportunity for the new generation to reinvent and rejuvenate the firm and restore the environment-organization fit that had often deteriorated pre-succession. We argue that CEO-related human capital enables successors to identify and amend misfits, allowing them to re-chart an organization’s course to enhanced performance, especially when the misfit is marked by a turnaround situation. Based on a quantitative analysis of 804 German firms and 31 qualitative interviews, we find support for these coherences and contribute to contingency and upper echelon theories.

KEYWORDS:

• CEO succession
• family firms
• upper echelon theory
• contingency theory
• mixed-methods

Introduction

In family firms CEOs typically and for good reasons exhibit particularly long – even epic – reigns (Gomez-Mejia et al., 2003; McConaughy, 2000; Stalk & Foley, 2012; Zona, 2016). At the same time, the general leadership literature argues that CEOs eventually become ineffective if they remain in office overly long as they are bound in their ability to re-tool their competencies at the same rate as the challenges that face their firms – which may result in a “misfit” between the firm’s current configuration and its challenges (Cannella et al., 2008; Miller, 1990, 1991, 1994). CEO succession, especially the post-succession phase, is thus seen as a context in which a new leader may act as a catalyst: she can assess the contemporary fit of the organization to current challenges – and may introduce refit actions accordingly (Miller, 1993).

The family firm succession literature classically centers on prominent topics of CEO successor selection (Ahrens et al., 2019; Calabrò et al., 2018) and important attributes of successors such as human capital (Chrisman et al., 1998; Le Breton-Miller & Miller, 2015), as well as prominence of familial relations and the succession process itself (Daspit et al., 2016; Handler, 1990; Le Breton–Miller et al., 2004). Yet, it is surprising how little is known about post-succession processes in family firms (Haddadj, 2003; Herriau & Touchais, 2015; Miller et al., 2003; Pardo-Del-Val, 2009) In particular, the strong focus on continuity and stability as central strategic elements of family firms (Miller & Le Breton-Miller, 2005) obscures our understanding of post-succession change actions along with their antecedents, prevalence, and most importantly performance impact. Further, the family firm context seems to disprove the general scholarly belief regarding the detrimental effects of long CEO reigns, given that family firms manage to prosper despite such long CEO tenures (Anderson & Reeb, 2003; Hansen & Block, 2020; Wagner et al., 2015). Hence, there are merits in investigating how family firms manage to utilize post-succession refit processes as a mechanism to address the inherent tensions between stability and the need to adapt to face current challenges and achieve high performance. Accordingly, the first research question that our study strives to answer is the following: “What is the performance impact of post-succession refit actions?” Beyond investigating the performance impact of these refit actions, our study also aims to answer the following two questions to improve our understanding regarding these crucial post-succession re-fit processes: “What are the important CEO-level antecedents for the realization of the refit actions?” and “How do these refit actions relate to performance in certain family firm succession contingencies such as turnaround situations?”

To address these questions, we invoke arguments from family firm literature as well as two well-established theories, Upper Echelon Theory (UET) and Contingency Theory (CT). Especially and when formulating predictions, we rely on the CT’s tenet of the importance of achieving a fit between the organization and its environment, as well as the hallmarks of UET that assign a central role to individual attributes of behavioral CEOs for imperfectly perceiving and interpreting an administrative setting, thereby for formulating and implementing strategic choices leading to firm performance (Cannella et al., 2008; Hambrick & Mason, 1984; Hambrick, 1983b). Indeed, family firm literature informs us about low rates of CEO turnover (Gomez-Mejia et al., 2003; McConaughy, 2000; Stalk & Foley, 2012; Zona, 2016) as well as often vivid family firm cultures that include strong relationships with rich social capital, rituals, and routines that reflect familial values – a setting which however is also known to create inertial forces and an emphasis on stability (Chirico & Nordqvist, 2010; Lansberg, 1983; Minichilli et al., 2014). While stability and management in the long-term are part of the family firm’s success formula (Miller & Le Breton-Miller, 2005), neglecting the inertial forces they may invoke would curb our understanding of how organizational change would emerge and affect family firm performance. With the arguments resting in CT, we show that this emphasis of family firms on stability and management in the long-term paradoxically also create a setting where, upon succession, there is often a suboptimal environment-organization fit. Moreover, we predict that this misfit is regularly large enough such that there is financial merit in overcoming inertia and in accommodating (a limited amount of) refit actions that leverage the optimization potential and restore environment-organization fit during family firm succession – that is, that such post-succession change usually has a (limited) positive relation to post-succession performance in many family firm successions. Moreover, we posit that the idiosyncratic cognitive and affective base of the successors affects whether they interpret their “construed reality” of the administrative situation (Hambrick & Mason, 1984; Sutton, 1987) in such ways that they become aware of this potential to improve environment-organization fit – which makes CEO attributes, such as CEO-related human capital, important antecedents of vital post-succession refit initiatives.

To test these conjectures, we harnessed data from standardized telephone interviews directly with 804 CEO successors in German family firms that cover 25 types of refit action implemented post-succession and that entail in-depth information on CEO attributes as well as post-succession performance. Using an industry- and trend-adjusted difference-in-difference approach to compare post-succession developments in profit margins, we found robust support for the above predictions and detail how succession contingencies, in particular successions under turnaround conditions that entail a high need for refit, affect them. Furthermore, we conducted 31 semi-structured open interviews with successors of similar family firms to complement our quantitative analysis by offering supporting statements for the arguments underlying our hypotheses.

The main contribution of our research is twofold. Firstly, it highlights the importance of change in increasing family firm performance upon succession. Our results show that change initiatives pertain to virtually all essential levels of the firm – from business relations to organizational structure and labor organization, from product to innovation and geographical activity. This finding is particularly interesting and counterintuitive given that it poses an exception to the success recipe of family firms focusing on continuity, stability, and long-term orientation (Miller et al., 2008). Accordingly, our results show that succession contexts offer family firms the opportunity to change, rejuvenate, and adapt. By enabling this process, change – contrary to common sense – helps to form the stable foundation on which the successful family firm organization of the next generations may rise. Our second contribution points out the importance of deliberate action by successors to identify and address the organization-environment misfit via refit actions (Cannella et al., 2008; van de Ven et al., 2013). Our findings show that CEO-related human capital is especially important and antecedes the performance-enhancing refit actions of successors. CEO related human capital enables successors not only to recognize the organization-environment misfit, but also to identify and implement refit actions to restore firm performance in the post-succession period (Donaldson, 1987; Hambrick & Mason, 1984). Beyond equipping them with the skills necessary to master the complex and turbulent nature of succession context (Daspit et al., 2016; Handler, 1990; Le Breton–Miller et al., 2004), human capital also helps successors as they balance the tensions between the necessity of renewal and the focus on stability inherent in family firms (Miller et al., 2008).

In a famous speech at the World Economic Forum in 1992, Edgar Schein compared organizational change to the challenge of entering the anxious atmosphere of a feared “Green Room”¹ – the unease that goes along with trying something new in a setting that treasures stability, with re-trying something in which the organization formerly failed, or simply the anxiety that occurs when a new leader formulates a new vision for a firm that entails cherished habits, routines and even workplaces being altered. If that is true, then our work contributes an overlooked but key insight into family firm research: we should perceive successful family firm successors as the “Heroes of the Green Room” – they regularly withstand this challenge and introduce vast amounts of change in spite of a setting that embraces the status quo with alluring arguments quite often grounded in the family rationale and tradition (Chirico & Nordqvist, 2010).² Moreover, our research unveils that these refit actions are the carriers of firm rejuvenation and innovation – the latter word in its original sense of “renewal” in Latin – and that they therefore entail a limited positive performance implication in the context of family firm successions. Thereby, our work facilitates a deeper understanding of why succession is such a critical phase for family firms. Its CT-UET perspective centers the role of adaptation and charts a course for a research agenda that is dedicated to preparing family CEO successors better for the important task of managing post-succession refit initiatives. Indeed, this theoretical angle on family firm succession highlights one of the most alluring dilemmas in family firm research: How CEO successors in family firms can preserve precious social capital, socioemotional wealth, and family values (Le Breton-Miller & Miller, 2015; Miller & Le Breton-Miller, 2005), but at the same time steer their firms through necessary refit actions and seduce their firms’ stakeholders to embrace “productive unlearning” processes that break with traditions, routines and old ways to the correct extent to be performant (Hedberg et al., 1976; Schein, 1993). Admittedly, our mixed-methods study may only be the tip of an iceberg, but it clearly reemphasizes the need for research on responsible and well-tailored adaptation in a family business setting (Cascio, 2002).

In this respect, by drawing on UET and CT we provide a theoretical perspective that enables a rich basis for an enhanced understanding of why those change processes that CEO successors initiate post-succession are often beneficial for firm performance in a family firm succession setting. This perspective makes it possible to see a seemingly inconspicuous and hidden aspect in the long-run and stability-oriented management approaches of family firms, namely, that suboptimal environment-organization configurations are likely to have silently (that is, often unnoticed by the predecessor) accrued prior to succession due to a combination of “stale-in-the-saddle” effects of seasoned predecessor CEOs at the firm’s helm (Miller, 1990, 1991, 1994) and the inertial forces particularly inherent to family firm (Chirico & Nordqvist, 2010). Taken together, for family firms this conceptually makes post-succession organizational adaption a necessary process in which misfits are amended as environment-organization fit is restored. Similarly, this theoretical view details important antecedents that are rooted in CEO-level attributes which affect her perception and evaluation of the firm’s situation in ways that are shielded from the inherent inertial forces of the firm (Hambrick, 2007) and that enable its performance potential to be leveraged through organizational refit actions. The moderated-mediation analyses of this paper lend further credibility to the above arguments, as it reveals how CEO successors with strong CEO-related human capital revitalize and reinvent family firms during succession, which is especially evident in a contingency when a misfit is aggravated as signaled by a turnaround situation (Chen & Hambrick, 2012; Donaldson, 1987). Finally, we contribute to the succession-change-performance nexus of UET and CT by adding contextualized research – or as Miller (1993, p. 656) put it: “Researchers might wish to examine the performance implications of succession in […] different organizational and environmental contexts, taking care to gauge the changes actually being made by the new leaders.” – for the most prevalent firm type across the world (Faccio & Lang, 2002; La Porta et al., 1999) – family firms.

Theory and hypotheses

Succession in family firms and successor ability

The importance of a successful transition to the next generation gives succession research an elevated place in the study of family firms (Sharma et al., 1996). A classic topic of family firm succession literature is the study of important attributes of successors that equip them to be the next generation of family firm leaders (Chrisman et al., 1998; Le Breton-Miller & Miller, 2015) capable of mastering the complexity of family firm succession as a multistage and multilevel process (Daspit et al., 2016; Handler, 1990; Le Breton–Miller et al., 2004). Moreover, in family firms CEO successions often occur in a context of unity of ownership and control that is held by a powerful incumbent, hence, ownership structure and agency issues (Demsetz & Lehn, 1985; Demsetz & Villalonga, 2001; Fama, 1980; Jensen & Meckling, 1976; Schulze et al., 2001) are reoccurring themes because they are known to affect family firm successions, and successor attributes specifically. In particular, those concentrated ownership and control structures that often occur in many family firms potentially reduce governance mechanisms over the incumbent CEO’s succession decisions to zero or only to internal checks, thus a potential pursuit of private benefit maximization may endanger firm performance around family succession (Dyck & Zingales, 2004; Johnson et al., 2000; Villalonga & Amit, 2006). For example, nepotism or gender preferences frequently lead to the appointment of CEO successors with inadequate attributes, such as a lack of CEO-related human capital and motivation, which result in negative performance effects (Bennedsen et al., 2007; Carnegie, 1889, 1933; Holtz-Eakin et al., 1993; Pérez-González, 2006; Smith & Amoako-Adu, 1999). These often outweigh the positive effects that a family background of the CEO and family continuity at the firm’s helm entail for post-succession firm performance (Ahrens et al., 2019).

While these embroilments around CEO ability (Daspit et al., 2016; Dawson & Parada, 2019; Hillebrand & Teichert, 2020; Löhde et al., 2020) and the succession process itself (Daspit et al., 2016; Handler, 1990; Le Breton–Miller et al., 2004) constitute well-trodden paths and central areas of attention in family firm research, less emphasis is generally placed on the years following succession, the so-called “post-succession” period, as well as the actual consequences of CEO succession in family firms from an organizational perspective. For instance, recent studies address the moral dilemma of the successor torn between family tradition and renewal upon succession (Erdogan et al., 2020; Radu-Lefebvre, 2021), others focus on particular initiatives such as the stimulating effect of a successful generational transition on innovation (Cabrera-Suárez et al., 2018; Carney et al., 2019; Hauck & Prügl, 2015; N. Zybura et al., 2021a, 2021b). However, the overall prevalence and scope of organizational adjustment processes in family firms, as well as their performance effect in the post-succession period, as well as the successor’s role in initiating them, deserves more scholarly scrutiny, at least in quantitative research (Hall et al., 2001).

Indeed, family firms typically emphasize stability, family continuity, and a long-run perspective in management (Miller & Le Breton-Miller, 2005). These are the building blocks of a strategy that enables them to nurture and maintain abundant social capital based on trust, reliability, continuity, and reciprocity that results in a competitive advantage (Arregle et al., 2007). These ingredients, however, are at first glance at odds with the idea of fluid change. Clearly, this prevents an intuitive inference, and draws a veil over the actual amount of change realized in the post-succession period as well as its performance implication in a family firm setting. Moreover, the complexity of this administrative setting already foreshadows that successor attributes are closely linked to both questions. Indeed, according to UET, managerial decision in the face of complexity is strongly affected by individual characteristics (Hambrick & Mason, 1984). Thus the aptness of the new CEO's decisions may depend on their CEO-related human capital (Adner & Helfat, 2003; Cannella & Rowe, 1995; Holcomb et al., 2009).

Drawing on contingency theory and upper echelons theory

In this article we rely on two theories – CT and UET – to create the theoretical perspective guiding our hypotheses. CT is rooted in the study of organization (March & Simon, 1958) and embraces the thought that there is no universal best way to lead and configure an organization (Lawrence & Lorsch, 1967) and, in order to maximize performance, the organizational design should rather fit the current situation the firm faces (van de Ven et al., 2013). This argument gives rise to one of the central tenets of CT: The concept of organization-environment fit that addresses the importance of achieving a fit between an organization and its environment and considers it to be one of the most important imperatives of the strategy process in organizations (Cannella et al., 2008; Hambrick, 1983a; Morgan, 1997). This is especially the case because the increasingly dynamic and drifting nature of today’s business environments may rapidly change the current situation with which a firm is confronted. Hence, fluidity in organizational strategy and design via strategic action to adapt to the needs of an ever-changing situation is seen as crucial to defending and maintaining organization-environment fit and, ultimately, to firm performance (Donaldson, 1987). Accordingly, leaders of organizations are obliged to screen their environment in order to identify changes, and modify their strategies to enable the organizational design elements such as organizational structure, firm culture, and choice of technology to match the needs of the changing environment (Volberda et al., 2012).

In this respect, it is fruitful to advance the theory and interlink CT with the central tenets of UET in order to understand what is to be expected in family firm successions in terms of change. UET shares similar roots in the study of organizations, but emphasizes the role of the CEO as an inherently constrained human decider whose perceptions and decisions, although intended to be rational, can only ever be rational within human boundaries (Cyert & March, 1963; March & Simon, 1958; March, 1978). Moreover, UET assigns a central role to the CEO as an important instance that affects an organization’s design including its strategic choices, and thereby firm performance (Hambrick & Mason, 1984). In particular, the theory posits that the CEO’s psychological, sociological, and experiential attributes that form her cognitive and affective base for decision making strongly affect the ways in which a CEO behaviorally perceives and interprets the administrative situation facing the firm (Hambrick, 2007; March & Simon, 1958). According to UET, the CEO’s highly individualistic ”construed reality” – which may differ markedly from the true reality – of the administrative situation (Sutton, 1987) will then partly predict strategic choices, and thus be reflected in organizational design and ultimately firm performance (Haleblian & Finkelstein, 1993).

Bringing those axioms of these two theories, the CT norm of organization-environment fit and the UET hallmark that strategic choice is bound by relatively stable CEO characteristics, into a coherent framework unveils key theoretical insights: a CEO may not re-tool her (usually relatively inert) personality or experiences at the same rate as the change in environment conditions as acknowledged by modern UET (Cannella et al., 2008). Hence, to her the fitting strategy, which would be optimal according to CT, might not be available or visible given their CEO characteristics. Thus, we can deduce that according to the above mechanisms such situationally misfitting CEOs can only achieve an inferior “boundedly rational” organization-environment fit that does not unleash the maximum possible performance.

Accordingly, this becomes such a central motive for CEO turnover and the installation of a new CEO whose attributes match (or fit) the firm’s current challenges (Cannella et al., 2008) that it brings about performance enhancing change, that is, refit actions, in order to re-achieve organization-environment fit (Donaldson, 1987). Vice versa, the above theoretical conjuncture implies that when a CEO succession only occurs inertly and upon generation change (as in many family firms), then a misfit-induced performance improvement potential has very likely accrued. Thus and primarily, this inertia at the CEO turnover level results in the UET and CT-driven prediction that in a family firm succession, productive change toward organization-environment fit is to be expected (CT), provided the successor has attributes which enable her to realize this improvement potential (UET).

Second and more generally, although the notion of achieving organization-environment fit via deliberate adaptive strategies is a reasonable one and has merit in explaining organizational success (Venkatraman & Prescott, 1990), general organizational inertial forces (different from those at the CEO turnover level) may prevent adaptation and result in misfitting and growing performance improvement potentials, perhaps even in demise (Hannan & Freeman, 1977). In fact, understanding the effects of organizational change on performance requires an assessment of the degree of general organizational inertia affecting the firm, because inertia limits the plannability, controllability, and seasonality of adaptive refit activities, and thus the resulting accrual of performance improvement potentials (Hannan & Freeman, 1984). In the next section, we apply and further contextualize the above theoretical conjuncture of CT and UET by focusing on identifying factors driving inertia in CEO turnover and at the organizational level in family firms to develop our hypotheses.

Inertial forces in family firms and post-succession refit actions

One of the most important aspects when contextualizing the above CT/UET framework is the insight that family firms are typically managed for the long-run (Miller & Le Breton-Miller, 2005) which results in the development of an internal equilibrium of well-established rules or norms, roles, and power distribution, and also of processes and routines in addition to an external equilibrium of economic and sociological embeddedness (Cyert & March, 1963; Granovetter, 1985; Nelson & Winter, 1982). While such a constellation may entail beneficial social capital that may drive performance (Ahrens et al., 2019; Arregle et al., 2007), this particular context paradoxically also harbors inertial forces because well-established structures and procedures, coupled with rich organizational history, experiences and habits, may also desensitize an organization to necessary adaptations (Hannan & Freeman, 1984; Hedberg et al., 1976; Miller, 1990, 1991, 1994; Nystrom & Starbuck, 1984; Schein, 1993). Indeed, experience often causes organizations and leaders to cling to old routines of behavior whilst – in the spirit of CT – optimal solutions to organizational challenges might have changed over time (Beck et al., 2008; Miller, 1990, 1991, 1993, 1994). In fact, many family firms are older organizations with a long tradition, and therefore are often particularly subject to inertial forces stemming from long-established structural and environmental embeddedness (Hannan & Freeman, 1977). Hence, it is possible that the adaptation of the stabilizing institutionalized fabric to a dynamic environment fuels opposition from organizational members (Coch & French, 1948), which makes reorganizing in such a context especially complex and potentially hazardous (Amburgey et al., 1990, 1993; Hannan & Freeman, 1984; Haveman, 1993). In addition to this, the firm culture in many family firms emphasizes the harmony between the family and the organization in order to shield embedded social capital, which also restricts the fluidity and flexibility of the firm to achieve a fit, especially if decisions involve actions that disturb trust, harmony, mutual obligations, or perennial family firm identity (Vincent Ponroy et al., 2019; Zellweger et al., 2010). In fact, research emphasizes the hesitation of family firms when it comes to important strategic decisions such as diversifying from their focus industry even if this is associated with economic losses (Gomez-Mejia et al., 2010). However, CT informs us that omitting such adaptative processes may result in a nonachieving organization-environment fit, hence the accrual of improvement potentials, and therefore preventing the firm from achieving its maximum performance (Donaldson, 1987).

In family firms, one factor aggravating this issue is inertia at the CEO turnover level. While normally in any given year one in six CEOs is subject to turnover and a CEO refit occurs at the helm of the firm (Kaplan, 2008), in family firms the CEO’s reign is connected to the family’s desire to retain control which makes CEO turnover a side aspect of a far less frequently occurring family generation change (Gomez-Mejia et al., 2003; Huybrechts et al., 2013; McConaughy, 2000; Stalk & Foley, 2012; Zona, 2016). Thus a CEO refit is seldomly seen in family firms and CEOs tend to remain in office for a very long time (Cannella et al., 2008; Miller, 1991), while upper echelons research informs us that CEOs remaining in office overly long become “stale” and inefficient (Miller, 1991). Further, upper echelons research highlights that successful CEOs who can look back on a lifelong career as the leader of a family firm, show a tendency to stick to formerly successful strategies and forms of organizations, and neglect to update them, even when facing new information, tougher competition or growing mismatches, just because of the alluring nature of past success to which research gives the name of the “Icarus paradox” (Miller, 1990, 1994). Moreover, in family firms a high level of exposure to established legacies, stories, and rituals (Jaskiewicz et al., 2015; Kammerlander et al., 2015) formed during this long period may curb the ability of a CEO to identify and address the possible opportunities (Ahrens et al., 2019). Thus, the context of family firms with its high inertia at the CEO turnover and organizational level is prone to lead to an unnoticed accumulation of improvement potentials due to a sub-optimal organization-environment fit upon family firm succession.

In this context, a CEO turnover would entail two distinct scenarios which are of analytical interest to our study.³ In the first of those, the CEO successor would not (be able to) identify and address the accumulated improvement potentials. This adherence to the current situation underlying the sub-optimal organization-environment fit would, in turn, result in no performance improvements. In the second scenario however, the CEO successor would (be able to) act as a necessary “reset,” assess the contemporary fit of the organizational equilibria and general business model, and initiate refit procedures. Just as tents are sometimes better than palaces (Hedberg et al., 1976), a new generation with fewer links to the corporate’s past experience may evaluate the firm’s configuration from a different perspective. The new CEO may have fewer commitments to past policies and organizational rules and is thus in a position to act as a catalyst for productive and adaptive change (Barker & Duhaime, 1997; Grinyer & Spender, 1979; Hofer, 1980; Miller, 1991, 1993; Tushman & Romanelli, 1985) and may spark a rejuvenating phase of organizational evolution, if not revolution (Greiner, 1972; Miller & Friesen, 1980). However painful those refits may be (Schein, 1993) and although change is not vital per se, we argue that in many circumstances these post-succession refit processes are vital for family firms, because it may give the firm the chance of a productive unlearning from its history and the chance to salvage those improvement potentials that accumulated due to inertia preventing a refit. We visualize this theory in Figure 1.

Figure 1. Pre-succession misfit, post-succession refit actions and performance.

To sum up, we argue that prior to many family firm successions a potential for improvements and adaptations has accumulated due to inertia that in this context is fueled by long-term power structures, established rules and procedures, an emphasis on history and experiences (Icarus paradox), stale-in-the-saddle effects, and anxious ignorance (Hannan & Freeman, 1984; Hedberg et al., 1976; Miller, 1990, 1991, 1994; Nystrom & Starbuck, 1984; Schein, 1993). In this context succession may offer a window of opportunity to overcome these barriers if the CEO successor chooses (or is able) to address the reasons for suboptimal organization-environment fit and spark a productive and adaptive refit process (Barker & Duhaime, 1997; Grinyer & Spender, 1979; Hofer, 1980; Miller, 1991, 1993; Tushman & Romanelli, 1985). By doing so, she can tap her company’s (limited) potential for improvements until refit is achieved, and thus translate this potential into enhanced firm performance:

H1:

In family firms, the amount of post-succession refit actions implemented by the CEO successor is log-positively related to post-succession firm performance.

The antecedent role of human capital in identifying and implementing refit actions

Family firm successions are complex (Le Breton–Miller et al., 2004) and, as we argued above, often go along with a nontrivial phase of organizational adaptation, potentially even organizational crisis (Miller, 1993). Thus, the decision making in such situations is demanding (Gladstein & Reilly, 1985; Smart & Vertinsky, 1977). UET accentuates the importance of a strong and apt cognitive base to ensure high decision quality especially when CEOs face complex tasks, (Hambrick & Mason, 1984), because the CEOs’ attributes affect how they behaviorally perceive and interpret the configuration of their firms (Hambrick, 2007; March & Simon, 1958). In line with this, behavioral scientists highlight the importance of human capital-related factors like education and experience within the cognitive base for decision outcomes (Morewedge, 2015; Ross, 1977). Namely, higher human capital of the decision maker facilitates a more rigorous mental representation of the organization and its environment which is less flawed by a variety of cognitive distortions, shortcuts, and biases that are applied in the face of overcomplexity (Baron, 2007; Cyert & March, 1963). Thus, CEO-related human capital, can lead to a more accurate assessment of the organizational aspirations that need to be satisfied for refit, and has therefore been identified by existing research as a key managerial attribute that drives family firm performance in the post-succession period (Ahrens et al., 2019; Simon, 1955). In line with this, we argue that the successor CEO-related human capital has an important antecedent role for post-succession refit actions. Specifically, we posit that successors with higher CEO-related human capital are more successful at identifying the misfit between the organization and its environment, and more comprehensive in detecting viable ways to address this misfit via implementing refit actions when compared to less able successors (Adner & Helfat, 2003; Cannella & Rowe, 1995; Holcomb et al., 2009; Pérez-González, 2006; Trow, 1961). Therefore, we hypothesize:

H2:

In family firms, CEO-related human capital of successors is positively related to the amount of post-succession refit actions implemented.

Turnaround situations as a moderator for refit actions

Arguably the extent of organization-environment misfit determines the extent of refit action necessary as well as the amount of viable refit actions that able CEO successor can discover and implement. For instance, firms which suffer from performance declines are more likely to have failed to adapt when compared to the firms without such a decline. Indeed, if a family firm encounters a turnaround situation, which is defined as a substantial crisis of “established firms that once performed satisfactorily, specifically in terms of profitability, but no longer do” Chen & Hambrick (2012, p. 225) during succession, then this pressurizes the firm to pursue efficiency enhancing adaptations (Harrigan, 1980; Whetten, 1987; Zammuto & Cameron, 1985) or to evade the downturn by pursuing a strategic reorientation (Barker & Duhaime, 1997). Hence, companies in such a weak position will have a higher need for refit action compared to companies that are strongly positioned (Barker & Duhaime, 1997). As a result, successions in turnaround situations are characterized by an urgent need for refit and respective strong performance improvement potential, hence more refit activity in the post-succession period.

This coherence makes the role of CEO-related human capital in identifying and addressing the improvement opportunities via refit actions even more crucial, in fact, leaders with higher human capital are found to be more equipped to manage such crisis situations (Wooten & James, 2008). Indeed, successions under turnaround situation conditions increase the general level of complexity and uncertainty associated with succession (Chen & Hambrick, 2012), because such unstable situations often cannot be dealt with by traditional management strategies (Comfort, 2007) and thus require particularly able leadership. Due to this elevated complexity, we argue that the positive relationship between CEO-related human capital and the implementation of refit actions in family firm succession (H2) is positively moderated, so that particularly able successors excel over unable successors when a complex and nontrivial organization-environment misfit has occurred, as is marked by a turnaround situation. Hence, we posit:

H3:

In family firms, the positive relationship between the CEO-related human capital of successors and amount of the post-succession refit actions implemented is positively moderated (that is, strengthened) by turnaround situations.

It directly follows from the above arguments that the positive relation between the human capital of the successor CEO and firm performance documented by prior research into family firm successions (Ahrens et al., 2019; Pérez-González, 2006) is likely to be mediated by the amount of refit actions implemented by the successor. Namely, post-succession refit actions serve as a tool in the repertoire of a successor with strong CEO-related human capital in order to salvage the performance improvement potentials rooted in an organization-environment misfit that had previously accrued before succession due to the typical inertial structures in family firms. Further, we argue that this indirect relationship will be conditional on the succession being characterized by a strong organization-environment misfit, as marked by a turnaround situation (see Figure 2 for the full framework). Therefore, we posit:

Figure 2. Conceptual diagram for hypotheses H1–H4.

^aH4: The relationship between human capital and firm performance is mediated by refit actions and moderated by a turnaround situation.

H4:

In family firms there is an indirect relationship between CEO-related human capital and firm performance, which is mediated by the amount of refit actions implemented by the CEO successor. This mediation is positively conditional on the succession being characterized by a turnaround situation.

Methods

Given the complex nature of family firm successions, we rely on a quantitative-dominant mixed-methods approach (Aguinis & Molina-Azorín, 2015; Jick, 1979; Molina-Azorin, 2012), where a complementary qualitative analysis in form of semi-structured open interviews with 31 CEO successors serves as a further perspective in order to deepen and enable a more nuanced understanding of quantitative findings. Data collection activities started and completed in 2010 for the gathering of quantitative data via standardized interviews whereas gathering of qualitative data via semi-structured open interviews which was initiated at the same time, lasted until 2021. Below, we detail quantitative and qualitative samples.

Sample for the quantitative analysis

The sampling procedure relies on the following pillars: the Mannheim Enterprise Panel (MUP), the Amadeus database, the Hoppenstedt database, the Creditreform solvency index information, German Bundesbank information, standardized computer aided telephone interviews (CATI), and web searches. CATI data provides succession details beyond the information typically available in public databases. Other data sources provide additional firm information. We begin by extracting a gross sample of German nonpublicly traded family firms from the MUP database and filter for the following criteria for the years 2002 to 2008: (1) 30 to 1,000 employees, (2) going concern, and (3) is a family firm. Thereby, we define the “family firm” attribute in the following way: a family firm is present if a maximum of three natural persons (as opposed to a legal entity or firm) own more than 50% of the enterprise and at least one of these owners is an executive director (Ahrens et al., 2019; Fiegener et al., 1994; Lansberg & Astrachan, 1994; Leach et al., 1990).

Employing a second filter on ownership and management details and changes, we single out 14,250 firms as “succession candidates” which are likely to have experienced a succession within this time horizon. We exclude firms without a telephone number (less than one percent) and the ISIC Rev. 3.1 sections (A) agriculture, hunting and forestry, (B) fishing, (C) mining, quarrying, (E) electricity, gas, and water supply, (L) public administration and defense and compulsory social security, (P) activities of households, (Q) extra-territorial organizations and bodies as well as division (91) activities of membership organizations. We collect financial data and impute missing data by employing the following cascade: (1) MUP, (2) Amadeus, (3) Creditreform, (4) Hoppenstedt, and (5) web searches. In a third step, “succession candidate” firm CEOs were contacted to participate in the CATI. Via a screening section, we ensure that we interviewed CEOs successor and ruled out interviewing non-succession firms. In particular, we required the confirmation of the following screening questions: (1) the interviewee is a successor, (2) the interviewee is an executive director, (3) the interviewee holds an ownership fraction of the firm or the transfer of ownership is planned, and (4) the succession took place between the years 2002 and 2008. As family firm succession comprises the transfer of management (Alcorn, 1982) and ownership (Barry, 1975) and since both may occur in different point in time we ask for the year when the successor became an executive director. The CATI covers detailed successor and succession information, in particular the post-succession managerial decision pattern of the CEO successor, and firm performance. Performance and firm size information from CATI is placed highest in the data hierarchy. Employing Bundesbank price index information, all values are reported in 07/2009 euros. We arrive at a sample of 804 CEO successions and a response rate of 29% of succession cases in this study that relies on difficult-to-obtain upper echelon data (Daily et al., 2003).

Variables

Our first dependent variable is profit margin (PM) which is a straightforward measure of firm performance frequently used by practitioners such as the executives participating in our CATI. More specifically, we collect data on performance in the succession year (first component) and the year 2009 (second component) to derive a differential measure of post-succession PM. Several upsides motivate this choice. First, a differential measure is advantageous because it controls for time-invariant unobserved heterogeneity by construction. Second, allowing the first component of the difference to randomly vary within the above succession event window (years 2002–2008) yields an average timespan of 3.5 years and importantly allows us to increase the number of comparable cases that we can interview sevenfold. The advantages of observing more cases while implementing controls for years in our specifications exceed the minor drawbacks that result from this random variation. Third, keeping the economic and environmental frame of the interview fixed and similar by CAT-interviewing in 2010 and referring to 2009 helps to reduce framing issues that can affect the CEOs’ answers in CATIs and would likely have occurred if we had interviewed over multiple years (Tversky & Kahnemann, 1981). Moreover, we follow advances in the performance measurement literature (Barber & Lyon, 1996), and apply industry-adjustments to PM to account for industry effects. Further, we account for mean reversion and the fact that firms with good initial conditions do follow other performance trends as compared to firms in a poorer shape by introducing performance adjustment (Barber & Lyon, 1996).⁴ The applied dependent variable Δ industry and performance-adjusted profit margin (PM) is thus cleansed of expected industry- and performance effects, hence it is a measure of “abnormal” performance (Pérez-González, 2006).

Our second dependent variable (and mediator/independent variable of interest) is the sum of refit actions. For this we inquire if the CEO successor reports post-succession adaptations in 25 business specific activities encompassing five domains, that is, in the CATI we elicit categorial responses with regard to whether the CEO successor implemented changes according to the categories within each of the five domains by directly asking for each item. These domains encompass (1) labor organization, (2) organizational structure, (3) product and innovativeness, (4) business relations, and (5) geographic activity (see Table 1 for the full list of these binary items). We then aggregate this information into an index variable sum of refit actions and standardize this variable for the ease of interpretation. By measuring broadly within five domains, we take a holistic approach. This choice was made because we assumed that the potential for improvement through refit at the critical juncture of family firm succession would be likely to take a very case-specific form, thus a broader measure would be more likely to capture organizational redesign activity (van de Ven et al., 2013). Moreover, we assumed that the successor would probably aim to implement those changes that offer the most refit potential in the case-specific local context, such that the sum of refit actions is likely to reflect this, although it may be composed by very different items in each specific case.⁵ The choice of the five domains rests on theoretical advances in the study of organizations (Kieser & Walgenbach, 2010). Specifically, we assumed that the case-specific form of the misfit could involve elements of internal organizing – like organizational activities, workforce composition, and organizational design – (for example, Nadler & Tushman, 1980) and could involve relationships between the organization and its environment – links to customers, suppliers, and so on. (Scott, 2004). Herein, the works by Kieser and Walgenbach (2010), Mintzberg (1979) and Porter (1985) were foundational for deriving the potential areas of change inside the organization. Similarly, the works by Scott (1995) and Powell (1990) that emphasize the embeddedness of organizations find echo in items concerning the external links and relationships that a CEO successor might want to change (Pfeffer & Salancik, 1978).

Table 1. Individual refit action categories, human capital and economic contingency.

	Economic contingency			Human Capital
Variable	Non-turnaround	Turnaround	Diff.	LHC	HHC	Diff.
Sum of refit actions	8.95	10.19	−1.24**	7.59	9.50	−1.90***
Labor organization (%)
New executive directors	51.49	51.35	0.13	43.77	53.22	−9.44***
Dropped executive directors	26.49	36.49	−10.00	20.20	28.16	−7.96***
Flattened hierarchy	29.00	27.03	1.97	22.64	31.46	−8.83***
Steepened hierarchy	12.25	14.86	−2.61	13.51	13.93	−0.42
Working time policy	32.43	41.89	−9.47	28.28	38.58	−10.30***
Compensation scheme	33.91	43.24	−9.33	28.28	38.80	−10.52***
Organizational structure (%)
Purchasing	48.64	64.86	−16.23***	42.57	53.33	−10.77***
Production	52.24	51.39	0.85	43.20	50.68	−7.48**
Marketing and sales	65.92	78.38	−12.46**	56.57	71.65	−15.09***
Personnel	52.72	48.65	4.07	44.97	53.66	−8.69***
Corporate finance and controlling	57.67	66.22	−8.54	45.97	59.33	−13.36***
Products and innovativeness (%)
Product innovation	22.83	18.92	3.91	20.20	25.33	−5.13*
Additional products	59.31	58.11	1.20	54.55	60.22	−5.68
Additional methods of production	54.48	51.35	3.13	46.10	52.65	−6.55*
Sorting out of products (heavy)	2.74	6.76	−4.01	3.05	3.12	−0.07
Sorting out of products (moderate)	28.93	43.24	−14.32**	25.42	33.41	−7.98**
Business relations (%)
New customers	85.89	87.67	−1.78	80.41	87.08	−6.67***
Loss of old customers	23.51	28.38	−4.86	21.28	25.89	−4.61*
New suppliers	54.98	64.86	−9.89	47.96	59.60	−11.64***
Dismissal of old suppliers	34.41	48.65	−14.23**	29.21	37.72	−8.51***
New bank relations	15.56	21.62	−6.07	13.09	18.40	−5.31**
New financiers	18.77	27.40	−8.63	11.90	20.13	−8.23***
Geographical activity (%)
Regional markets	−2.22	−6.76	4.53	−1.02	−3.77	2.75
National markets	2.47	4.05	−1.58	3.05	2.66	0.39
International markets	6.67	8.11	−1.44	1.02	7.76	−6.74***

Notes: The t-test results are indicated in the difference column. Significance levels: ***p < .01, **p < .05, *p < .1.

We measure CEO-related human capital (the second independent variable of interest) using the human capital score (HCS) of (Ahrens et al., 2015). This proxy score is composed of the sum of five indicator variables (CATI data) which include (1) age (years) of the successor is above median of the successors (proxy for general experience), (2) industry experience (years) is above median of the successors (proxy for industry experience), (3) a leadership experience indicator (proxy for practical managerial skills), (4) business education, an indicator if the successor holds a university degree in business studies (or strongly related field) or was educated at a university of cooperative education (proxy for theoretical managerial skills), and (5) use of a business plan during the succession (proxy for professional managerial skills).

To capture a context with a marked need for refit actions, we utilize a proxy for a succession occurring under turnaround situation conditions. The binary variable turnaround takes the value 1 if the company was suffering both from a low PM (at least 1% lower than the average industry PM at the two digit ISIC level) during an industry downturn (average industry PM lower than 2%), or if its PM was lower than 0.5% in the succession year (Ahrens et al., 2019).

We employ a comprehensive vector of controls. This includes succession year firm size to control for size effects, as well as standardized corporate age and corporate age² to control for the nonlinear coherence of age-performance relationship (Coad et al., 2018). To capture particular effects of founder and post-founder generations we utilized two indicators: generation one which is equal to 1 if corporate age is less or equal to 25; and generation two which is equal to 1 if corporate age is higher than generation one but less or equal to 50 (Adams et al., 2009; Anderson & Reeb, 2003). Similarly, we control for unplanned successions via a binary variable that signals succession due to death or disease, as the absence of planning affects performance (Hillier & McColgan, 2009; Le Breton–Miller et al., 2004; Slovin & Sushka, 1993). To capture potential heir apparent effects as well as the effects ownership can have on performance (Behn et al., 2005), we take into account if the successor held ownership share prior to management transfer (ownership > leadership transition), as well as the number of years the successor held ownership upon succession (successor ownership in years). Moreover, we control for successor types, for example, for external successor (that is, if the successor did not work at the firm prior to succession), for non-family insider successor (that is, if the successor was not a family member but was working at the firm prior to succession), and for imprinted successor (that is, if the successor spent more years at the firm than the average of insider successors), since each of those statuses might affect the initiation of refit processes and affect performance (Zhang & Rajagopalan, 2010). We also control for previous owner involvement after the succession takes place given this can curb the discretion of the successor (Ahrens et al., 2018; Mussolino & Calabrò, 2014). Finally, the sudden financing requirements variable indicates if unexpected financing requirements occur in the first year after succession, as this may affect the leeway of the successor to initiate refit, and thus performance. To account for remaining effects of mean reversion, we include the industry adjusted PM as well as industry and performance-adjusted PM in the succession year (Ahrens et al., 2019; Pérez-González, 2006). Finally, we include in all our models years since succession, region (Central, North, and South Germany and left East Germany as base), legal form controls (stock corporation, sole proprietorship, and left other legal forms as base). Summary statistics and pairwise correlations of regression variables are presented in Table 2.

Table 2. Summary statistics and pairwise correlations.

Nr.	Variable	Mean	SD	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
1	Δ Industry and performance-adjusted PM	1.316	4.464	1.000
2	Sum of refit actions (log)	0.000	1.000	0.095*	1.000
3	Human capital score	2.801	1.221	0.076	0.209***	1.000
4	Turnaround	0.154	0.362	−0.066	0.084*	0.098**	1.000
5	Industry and performance-adjusted PM	0.171	1.769	−0.135***	0.077	−0.065	−0.064	1.000
6	Industry-adjusted PM	2.817	5.857	0.052	−0.023	−0.182***	−0.367***	0.518***	1.000
7	Number of employees	80.147	82.083	−0.058	0.103***	0.054	−0.029	−0.027	−0.109**	1.000
8	Corporate age	0.000	1.000	−0.058	0.095**	−0.038	0.093**	−0.101**	−0.113**	0.045	1.000
9	Years since succession = 1	0.100	0.300	0.042	0.046	0.094**	−0.022	−0.022	0.042	−0.040	0.118***	1.000
10	Years since succession = 2	0.101	0.301	0.079*	0.076**	0.055	−0.002	0.006	0.056	−0.025	0.041	−0.111***	1.000
11	Years since succession = 3	0.139	0.346	0.008	0.039	0.069*	0.107**	−0.107**	−0.053	−0.045	0.052	−0.134***	−0.135***	1.000
12	Years since succession = 4	0.155	0.363	−0.108**	0.023	−0.043	0.014	0.030	−0.000	−0.004	−0.007	−0.143***	−0.144***	−0.173***	1.000
13	Years since succession = 5	0.133	0.340	0.076	−0.026	−0.045	−0.070	−0.000	0.041	0.014	−0.039	−0.130***	−0.131***	−0.158***	−0.168***	1.000
14	Years since succession = 6	0.150	0.358	0.084*	−0.045	−0.069*	−0.009	−0.026	−0.060	0.007	−0.059	−0.140***	−0.141***	−0.169***	−0.181***	−0.165***	1.000
15	Years since succession = 7	0.221	0.415	−0.126***	−0.081**	−0.031	−0.017	0.087*	−0.003	0.071**	−0.068*	−0.177***	−0.178***	−0.215***	−0.229***	−0.209***	−0.224***	1.000
16	East Germany	0.139	0.346	0.032	0.023	0.009	−0.023	0.009	−0.026	−0.024	−0.085**	−0.050	−0.015	−0.027	0.006	0.043	0.062*	−0.024
17	North Germany	0.197	0.398	−0.004	0.024	0.028	−0.037	−0.024	−0.003	−0.006	−0.018	0.034	−0.062*	−0.036	0.004	−0.009	0.019	0.038
18	Central Germany	0.335	0.472	0.034	−0.003	0.026	0.065	0.051	−0.038	0.002	−0.028	−0.024	0.087**	0.004	−0.028	0.017	−0.004	−0.035
19	South Germany	0.330	0.470	−0.056	−0.034	−0.056	−0.017	−0.038	0.061	0.020	0.105***	0.032	−0.024	0.046	0.020	−0.041	−0.058*	0.021
20	Stock corporation	0.036	0.187	0.074	−0.022	0.015	−0.065	0.019	0.025	−0.037	0.081**	0.002	0.047	0.057	0.009	−0.016	0.012	−0.087**
21	Sole proprietorship	0.002	0.050	0.009	−0.024	0.008	−0.028	−0.009	0.062	0.006	0.040	−0.017	−0.017	−0.020	0.047	−0.020	0.049	−0.027
22	Other legal forms	0.961	0.193	−0.071	0.027	−0.017	0.071	−0.014	−0.047	0.035	−0.089**	0.002	−0.041	−0.050	−0.021	0.021	−0.024	0.092***
23	Generation one	0.383	0.487	0.121**	0.050	0.017	−0.020	0.100**	0.037	−0.020	−0.545***	−0.010	−0.037	−0.069*	−0.023	0.021	0.060*	0.042
24	Generation two	0.332	0.471	−0.056	−0.144***	0.012	−0.101**	−0.019	0.045	−0.019	−0.162***	−0.048	−0.020	0.012	0.076**	−0.031	−0.022	0.017
25	Unplanned succession	0.091	0.288	0.096**	0.079**	0.052	0.080*	0.012	−0.008	−0.025	0.006	0.025	0.009	−0.003	0.031	0.003	0.001	−0.053
26	Successor ownership in years	5.986	5.618	0.004	−0.030	0.019	−0.015	−0.082*	−0.005	−0.023	0.056	0.141***	0.091***	0.112***	0.041	−0.006	−0.136***	−0.175***
27	Ownership > leadership transition	0.690	0.463	0.015	−0.087**	−0.036	−0.017	−0.051	−0.016	−0.065*	0.008	−0.002	0.037	−0.018	0.028	0.025	−0.019	−0.038
28	Imprinted successor	0.365	0.482	−0.074	−0.151***	0.156***	−0.038	−0.038	−0.047	−0.075**	−0.011	0.035	0.004	−0.068*	−0.041	0.097***	0.031	−0.040
29	External successor	0.179	0.384	0.055	0.131***	0.238***	0.020	−0.028	−0.082*	0.116***	−0.064*	0.007	0.048	0.009	−0.030	−0.097***	−0.033	0.087**
30	Non-family insider successor	0.198	0.399	0.008	−0.023	0.107***	−0.021	−0.036	−0.055	−0.048	−0.094***	−0.019	−0.011	−0.110***	0.037	0.017	0.036	0.036
31	Previous owner involvement	0.633	0.482	−0.042	−0.042	−0.085**	−0.097**	0.091**	0.031	0.016	−0.130***	−0.014	−0.131***	−0.025	−0.065*	0.002	0.065*	0.126***
32	Sudden financing requirements	0.104	0.305	0.026	0.190***	0.065*	0.150***	−0.017	−0.065	−0.002	0.075**	0.064*	0.063*	0.064*	−0.045	−0.025	−0.003	−0.083**
Nr.	Variable	16	17	18	19	20	21	22	23	24	25	26	27	28	29	30	31	32
16	East Germany	1.000
17	North Germany	−0.199***	1.000
18	Central Germany	−0.285***	−0.351***	1.000
19	South Germany	−0.282***	−0.347***	−0.497***	1.000
20	Stock corporation	−0.020	0.005	−0.024	0.035	1.000
21	Sole proprietorship	0.052	−0.025	0.018	−0.035	−0.010	1.000
22	Other legal forms	0.006	0.002	0.019	−0.025	−0.966***	−0.249***	1.000
23	Generation one	0.272***	−0.007	−0.021	−0.173***	−0.029	−0.040	0.010	1.000
24	Generation two	−0.220***	0.026	0.010	0.129***	−0.025	−0.036	0.026	−0.556***	1.000
25	Unplanned succession	−0.014	−0.057	0.078**	−0.020	0.031	−0.016	−0.062*	0.042	−0.056	1.000
26	Successor ownership in years	−0.065*	−0.056	0.031	0.064*	0.007	−0.013	−0.013	−0.081**	0.059*	0.029	1.000
27	Ownership > leadership transition	0.013	−0.048	0.007	0.023	0.058	−0.020	−0.061*	0.042	−0.041	0.081**	0.506***	1.000
28	Imprinted successor	−0.045	0.005	0.001	0.028	0.052	0.014	−0.058	−0.111***	0.136***	0.027	0.098***	0.047	1.000
29	External successor	0.018	−0.011	0.074**	−0.079**	0.014	−0.023	−0.007	0.169***	−0.121***	−0.034	−0.096***	−0.024	−0.364***	1.000
30	Non-family insider successor	0.080**	0.022	−0.001	−0.076**	−0.046	0.038	0.033	0.103***	−0.037	0.083**	−0.091***	−0.039	0.198***	−0.232***	1.000
31	Previous owner involvement	−0.007	0.088**	−0.046	−0.023	−0.131***	−0.014	0.113***	0.104***	0.033	−0.280***	−0.099***	−0.218***	0.076**	−0.092***	0.012	1.000
32	Sudden financing requirements	−0.018	0.028	0.019	−0.030	0.087**	−0.017	−0.085**	0.041	−0.102***	0.034	−0.018	−0.019	−0.023	0.097***	−0.014	−0.059*	1.000

Notes: Pearson correlation coefficients offered as the measure for the pairwise correlations. Significance levels: ***p < .01, **p < .05, *p < .1.

Quantitative empirical strategy

To test Hypotheses 1 to 3 we rely on OLS estimations and Huber–White robust standard errors. For the moderated mediation hypothesis (H4), we additionally harness conditional process modeling (Hayes, 2018). For this purpose, we use two regression specifications (Models 5 and 6, Table 3) to estimate the path coefficients. These regression specifications, which encompass the moderated mediation paths visualized in Figure 2 and are described in model equations below respectively, yield the parameters required to calculate the magnitude of the moderated mediation effect.

$\left(\mathit{M}5\right)\mathit{RE}\mathbf{-}\mathit{FIT}\mathit{ACTIONS}={\gamma }_5+a_5\times \mathit{HCS}+b_5\times \mathit{TURNAROUND}+c_5\times \mathit{TURNAROUND}\times \mathit{HCS}+{\sum }_{i=1}^N\left(z_{5\mathit{i}}\times \mathit{CONTRO}{L}_i\right)$

$\left(\mathit{M}6\right)\widehat{\mathbf{PERFORMANCE}}={\gamma }_6+a_6\times \mathit{HCS}+b_6\times \mathit{TURNAROUND}+c_6\times \mathit{TURNAROUND}\times \mathit{HCS}+d_6\times \mathit{RE}-\mathit{FITACTIONS}+{\sum }_{i=1}^N\left(z_{6\mathit{i}}\times \mathit{CONTRO}{L}_i\right)$

Table 3. Regression analysis.

	DV: Δ Industry and performance-adjusted PM		DV: Sum of refit actions (log)			DV: Δ Ind. and perf.-adj. PM
	Base I	H1	Base II	H2	H3 & H4(I)	H4(II)
Variables	(1)	(2)	(3)	(4)	(5)	(6)
Sum of refit actions (log)		0.62**				0.47*
		(0.254)				(0.257)
Human capital score				0.18***	0.15***	0.54**
				(0.047)	(0.051)	(0.234)
Turnaround					−0.75†	1.13
					(0.488)	(1.794)
Human capital score and turnaround					0.23*	−0.49
					(0.133)	(0.535)
Industry and performance-adjusted PM	−0.57**	−0.64***	0.06**	0.04†	0.04†	−0.72***
	(0.224)	(0.230)	(0.029)	(0.027)	(0.028)	(0.244)
Industry-adjusted PM	0.11**	0.12**	−0.01	−0.00	−0.00	0.13**
	(0.055)	(0.056)	(0.010)	(0.009)	(0.010)	(0.063)
Number of employees	−0.00†	−0.00*	0.00***	0.00***	0.00***	−0.00†
	(0.001)	(0.001)	(0.000)	(0.000)	(0.000)	(0.001)
Corporate age	0.50	0.90	−0.02	−0.11	−0.13	0.77
	(0.609)	(0.639)	(0.128)	(0.132)	(0.132)	(0.663)
Corporate age^2	−0.03	−0.07	0.01	0.02	0.02	−0.05
	(0.070)	(0.072)	(0.015)	(0.015)	(0.015)	(0.074)
Generation one	2.19*	2.89**	−0.09	−0.21	−0.25	2.76**
	(1.157)	(1.245)	(0.239)	(0.235)	(0.235)	(1.306)
Generation two	0.99	1.69*	−0.29†	−0.44**	−0.47**	1.46
	(0.905)	(0.981)	(0.198)	(0.197)	(0.197)	(1.040)
Unplanned succession	1.41*	1.13†	0.21	0.13	0.15	1.24†
	(0.753)	(0.773)	(0.178)	(0.179)	(0.176)	(0.769)
Successor ownership in years	0.01	0.01	−0.01	−0.00	−0.00	0.03
	(0.039)	(0.040)	(0.009)	(0.009)	(0.009)	(0.040)
Ownership > leadership transition	−0.35	−0.35	0.09	0.03	0.04	−0.47
	(0.553)	(0.575)	(0.112)	(0.115)	(0.117)	(0.583)
Imprinted successor	−0.81*	−0.78†	−0.15	−0.26**	−0.24*	−1.23**
	(0.477)	(0.502)	(0.115)	(0.126)	(0.129)	(0.531)
External successor	0.20	0.15	0.04	−0.09	−0.10	−0.48
	(0.616)	(0.646)	(0.112)	(0.124)	(0.126)	(0.709)
Non-family insider successor	0.01	0.29	−0.13	−0.22*	−0.23*	−0.06
	(0.534)	(0.518)	(0.118)	(0.120)	(0.123)	(0.551)
Previous owner involvement	−0.14	−0.28	−0.08	−0.03	−0.02	−0.36
	(0.486)	(0.497)	(0.104)	(0.106)	(0.107)	(0.506)
Sudden financing requirements	−0.12	−0.31	0.48***	0.47***	0.45***	−0.02
	(0.842)	(0.854)	(0.106)	(0.111)	(0.114)	(0.862)
Year, region, and legal form controls	✓	✓	✓	✓	✓	✓
Constant	0.54	−0.27	0.28	−0.26	−0.13	−1.91
	(1.339)	(1.383)	(0.252)	(0.310)	(0.313)	(1.626)
Observations	409	385	414	392	392	370
R-squared	0.12	0.14	0.13	0.16	0.17	0.16
F test	2.65	2.61	4.04	4.02	3.82	2.77
Prob F	0.00	0.00	0.00	0.00	0.00	0.00

Notes: Robust standard errors are in parentheses. Significance levels: ***p < .01, **p < .05, *p < .1, †p < .15.

From the results of these specifications, the coefficient of the indirect effect of the hypothesized moderated mediation is calculated as $\mathit{\hspace{0.17em}}\left(a_5+\mathit{\hspace{0.17em}}{c}_5\times \mathit{TURNAROUND}\right)\times d_6$ (Hayes, 2018). As the effect size is conditioned on whether the succession has a marked need for refit actions, it is additionally probed with the realizations of the turnaround variable (0 or 1 respectively) to investigate differential effect sizes and thus test for a moderation of the mediation effect. We then harness the bootstrapping estimation method with 5000 repetitions to derive bias corrected standard errors for respective coefficients and calculate the index of the moderated ($c_5\times d_6)$ mediation, whose statistical significance would indicate a moderated mediation relationship (Hayes, 2018).

Complementary qualitative analysis

Our complementary qualitative analysis rests on recordings of personal interviews conducted by the authors with 31 successors in family firms that match the firm characteristics of our quantitative sample in the sense that they are also typical medium-sized (in terms of employees) German family firms that have experienced a succession. However, to improve the external validity and ensure that artifacts of the quantitative analysis do not interfere with the qualitative analysis, the firms interviewed were not part of the quantitative sample and were not selected on the basis of the MUP database, but on the network of the authors’ institute. We chose a qualitative analytical and interpretive multiple case study methodology and rely on semi-structured open interviews to reflect the insights of the interviewees (Eisenhardt, 1989; Yin, 2012). Recorded interviews average 36 minutes, totaling 1112 min in length. Table 4 provides a contextual overview for each case.

Table 4. Contextual overview of the complementary qualitative analysis.

	Firm Characteristics			Successor Specific Characteristics						Succession Specific Characteristics
ID	Industry	Size (#Emp.)	Founding Period	Age Group	Non-family Successor	Prior Industry Experience	Prior Leadership Experience	Focus and Degree of Education	Preparedness	Turnaround Situation	Amount of Refit Actions Implemented	Planned Succession
1	Printing	151-500	1946–1980	36–45	No	No	No	Technical, non-academic	Prepared & groomed	No	High	Yes
2	Construction	50-150	1933–1945	18–35	No	No	No	Hybrid, academic	Prepared & groomed	No	Medium	Yes
3	Metal casting	50-150	1871–1918	18–35	No	Yes	No	Merchant, academic	Prepared & groomed	No	Medium	Yes
4	Automotive	50-150	1946–1980	36–45	No	Yes	Yes	Merchant, non-academic	Unprepared	No	Medium	No
5	Manufacturing	<50	1946–1980	46–55	No	No	Yes	Hybrid, academic	Prepared & groomed	No	Medium	Yes
6	Construction	151-500	1919–1932	46–55	No	Yes	Yes	Merchant, academic	Prepared & groomed	No	High	Yes
7	Metalworking	<50	1946–1980	46–55	Yes	Yes	Yes	Technical, academic	Prepared	No	Low	Yes
8	Lighting	151-500	1871–1918	36–45	No	No	No	Merchant, academic	Unprepared	Yes	High	No
9	Logistics	>500	1946–1980	36–45	No	No	No	Merchant, academic	Unprepared	No	Medium	Yes
10	Packaging	50–150	1400–1700	46–55	No	No	No	Hybrid, academic	Prepared & groomed	No	Medium	Yes
11	Packaging	50–150	1400–1700	36–45	Yes	No	Yes	Hybrid, academic	Prepared	No	Medium	Yes
12	Automotive	151–500	1946–1980	18–35	No	No	No	Technical, academic	Prepared	No	None	No
13	Manufacturing	<50	1981–2000	46–55	Yes	Yes	Yes	Technical, non-academic	Prepared	Yes	Low	Yes
14	Heating/Sanitary	50–150	1933–1945	36–45	Yes	No	Yes	Merchant, academic	Prepared	Yes	High	No
15	Electronics	151–500	1871–1918	18–35	No	No	No	Merchant, academic	Prepared & groomed	No	Medium	Yes
16	Packaging	151–500	1919–1932	36–45	No	No	Yes	Merchant, non-academic	Prepared & groomed	No	Low	Yes
17	Plastics	151–500	1946–1980	36–45	No	Yes	Yes	Merchant, non-academic	Prepared & groomed	No	High	Yes
18	Automotive	>500	1946–1980	18–35	No	Yes	No	Hybrid, non-academic	Prepared & groomed	Yes	High	Yes
19	Metalworking	50–150	1981–2000	36–45	Yes	Yes	Yes	Technical, non-academic	Prepared	No	Medium	Yes
20	Metalworking	50–150	1981–2000	18–35	Yes	No	No	Merchant, non-academic	Prepared	No	Medium	Yes
21	Signage	<50	1871–1918	56–65	Yes	Yes	Yes	Merchant, Non-Academic	Prepared & Groomed	No	Medium	Yes
22	Food Industry	50–150	1946–1980	36–45	Yes	Yes	Yes	Merchant, non-academic	Prepared	Yes	High	Yes
23	Furniture	151–500	1946–1980	36–45	No	No	No	Merchant, academic	Prepared & groomed	No	Medium	Yes
24	Cosmetics	>500	1946–1980	36–45	No	No	No	Merchant, academic	Unprepared	Yes	High	No
25	Lubricants	151–500	1981–2000	18–35	No	No	No	Merchant, academic	Prepared	No	Medium	Yes
26	Packaging	151–500	1919–1932	36–45	No	No	No	Merchant, academic	Prepared & groomed	Yes	High	Yes
27	Consulting	<50	1981–2000	>65	Yes	Yes	Yes	Merchant, academic	Prepared	No	Low	Yes
28	Manufacturing	>500	1400–1700	36–45	No	No	Yes	Merchant, academic	Prepared	No	None	Yes
29	Food Industry	<50	1981–2000	36–45	No	Yes	No	Merchant, non-academic	Prepared & groomed	No	High	Yes
30	Retail	<50	1400–1700	18–35	No	No	No	Hybrid, academic	Prepared & groomed	No	Low	Yes
31	Retail	50–150	1919–1932	36–45	No	No	No	Merchant, academic	Prepared & groomed	No	High	Yes

Our qualitative analysis followed the idea of the two-cycle coding approach of Saldaña (2013). After transcription, we started our qualitative data analysis with the first cycle. This entailed a broad descriptive coding (Saldaña, 2013) targeted at finding parts of the interviews that related to change upon succession which we then used to gain a general understanding of each succession case. We then moved on to our second cycle where we applied elaborative coding (Saldaña, 2013) to relate our descriptive codes to the hypotheses from our quantitative study. This allowed us to see if the assumptions of the theory that underly our hypotheses of the quantitative study (that is, there is a suboptimal fit before succession; the ability of the CEO is crucial to refit; refit actions have positive effects; “turnaround” is an amplifier for this) are – at least in traces – also echoed in the quotes of the CEO successors. Finally, in line with a quantitative dominant approach (Aguinis & Molina-Azorín, 2015), we report qualitative insights and excerpts that offer complementary in-depth insights into arguments and our quantitative findings (Greene et al., 1989).

Results

Quantitative analysis and hypothesis testing

An introductory descriptive analysis of each refit action reveals that refit action in the post-succession period is frequent across the entire firm, and more frequent in those successions where a successor with high CEO-related human capital takes over under turnaround conditions. Regression models for hypothesis testing are presented in Table 3.

Model 1 and Model 3 in Table 3 are control only models for both our dependent variables. In Model 2 we observe a positively significant coefficient (β = 0.619, p = .015) of the sum of refit actions (log) indicating a log positive performance relationship in line with H1 (Table 3, Model 2). Second, the coefficient of human capital score variable is significantly positive (β = 0.180, p = .000) when the sum of refit actions is the dependent variable (Table 3, Model 4, H2). Further, in Model 5 we observe this effect is strengthened if the succession is characterized by a high need for refit actions, more precisely the interactive combination of human capital score and turnaround is positively significant (β = 0.227, p = .090, H3). In Figure 3, we also provide the interaction plot illustrating the nature of this moderation relationship.

Figure 3. Interaction plot for human capital score and turnaround (Model 5, Table 3).

Results regarding the moderated mediation hypothesis (H4) are reported in Table 5. A positively significant index of moderated mediation (0.107) and a 95% confidence interval (95%CI: [0.042; 0.424]) suggest that the mediation of refit actions between successor human capital and firm performance differs significantly for turnaround and non-turnaround cases, hence it offers support for our H4. This support is also underlined by the higher mediation effect size (0.175; 95%CI: [0.094; 0.537]) for turnaround cases in comparison to the mediation effect size (0.068; 95%CI: [0.009; 0.221]) for non-turnaround cases. Furthermore, significance of the mediation effect for non-turnaround cases suggests that there is also a mediation of refit actions between human capital and firm performance in non-turnaround cases, but in a turnaround case this mediation is more than 1.5 times stronger.

Table 5. Bias corrected bootstrapped indirect effects (5000 repetitions).

	Coefficient	95%CI
Turnaround = 0	0.068	0.009	0.221
Turnaround = 1	0.175	0.094	0.537
Index of moderated mediation	0.107	0.042	0.424

Notes: The bootstrap command of STATA 14 is used to create the bootstrapped standard errors.

Quantitative post hoc analysis

In post hoc analyses, we reconsidered the breadth of our refit measure and re-ran our specifications with a narrower post hoc measure of the sum of refit actions. Specifically, we required that the sum is composed only of those items that entail a positive relation with performance in our data. We report that when harnessing this narrower “ex post” measure, the results for all paths remain robust. Moreover, we investigated the performance relationship of individual refit actions and report especially noteworthy findings here. In particular, while most of the refit actions are insignificant when analyzed in isolation, we observe that a moderate review of the product portfolio and an accordant moderate sorting out (β = 0.911, p = .050) is positively related to abnormal performance. Also, when combining the moderate and strong portfolio review indicator variables to a scale, we observe an inverted u-shape relationship with performance: a positive coefficient of sorting out of products (β = 2.357, p = .039) and a negative and significant coefficient of sorting out of products squared (β = −1.542, p = .075) which arguably indicates that not all that was seemingly “old” deserves to be abolished and that refit processes deserve diligence.

Complementary qualitative analysis

The semi-structured open interviews with CEO successors provide further detail beyond the above regressions on the nature and existence of suboptimal organization-environment fit upon generation change in family firms. Important and analogue themes such as post-succession refit actions due to accumulated improvement potentials, the value of CEO-related human capital for managing and designing adaptations, and the general merit of a limited refit post-succession similarly become visible. Due to the ancillary role of our complementary qualitative analyses, we present a short selection of powerful quotes in the text as well as further “proof quotes” in Table 6 (Pratt, 2008). In particular, the statements of the interviewed successors provide case-based evidence for, as well as color to, those suboptimal environment-organization configurations that are likely to have silently accrued upon succession (14) and were often unnoticed by seasoned incumbents (18) (see Table 6, Theme 1):

Table 6. Exemplary quotes from the qualitative analysis.

Theme	Exemplary Quote
Theme 1: Perceived suboptimal environment – organizational fit at succession	(14): “We used to have three tax and accounting consultancies, one for our assets and the wage payments, the next for the taxes and the third for public accounting. And still we factorized our bills while the payroll accounting was calculated by hand. […] The internal organization was really 30 to 40 years old, and it was about time to overhaul it. […] There was no way to communicate with the suppliers by e-mail or to communicate at all. The project leaders couldn’t write any letters themselves, so they dictated their bills on the dictaphone and then gave them to the typist. That’s not even 5 years ago. And the most crucial employee here at the company, was a lady who was able to write letters faster than the rest of the workforce. It might be funny, but it was actually the case. And the bottleneck was always when the lady was on vacation, no bills could be written and then there was no payment for 4 weeks and the liquidity was correspondingly bad during this time.” (18): “We introduced a quality assurance department. It wasn’t there before, although it’s very important. […]. That doesn’t mean that he [the predecessor] didn’t deliver quality, but just that you can double check and times have changed. If you’ve been in a company for a long time, you might become a little blind to the company.” (14): “Today […], it just doesn’t make sense here in [the small city], where you don’t have any walk-in customers, to run a store […] and bill €7.50 on average. That just didn’t make sense.” (26): “So, basically, it’s a family firm, and my dad was running the company for more than 30 years, but eventually […] he decided that he wanted to […] retire and pass on the business. [Upon succession] We didn’t have any structures that would allow for growth, we didn’t have any decision-making processes that would enable other people, except for my dad, to make a decision and to move on with whatever. So, we had to start discussing what decisions could be transferred to other people, what were the competencies, how far can they go.” (14): “The predecessors also knew that the workforce was too large. They just didn’t know where; is it the fitter, or the cleaning lady, or the part-time worker? It was very difficult for them to find out. […] But if you’ve been in the company long enough, you slowly get to wear blinkers over time. […]. One problem before was certainly to implement things, even if you saw them. The predecessor had a hard time with that. And then one of my jobs was to sit down with the people at a table and say: ‘Sorry, we have to restructure and that position will be cut.’ That was the biggest obstacle, of course. That was of course a problem in [the small town], because the company hadn’t attracted attention for the last 75 years by laying people off, but by growing continuously.” (22): ”[Situation upon succession] What didn’t work was that the company wasn’t structured at all even with 40–50 people, and if you triple that within 3 years then you simply have chaos. That did not work. […] Until last year, we did not even have a sales and operation plan.” (9): “We have become more international. […] We didn’t have that before. […] this authoritarian management style […] ‘I decide everything because I know everything,’ which I think was still possible in the old times […] that has fundamentally changed. My management style is not authoritarian.” (21): “With the [predecessor], the employees had to be there from then until then […]. Then I have my ‘sheep’ around me, so clearly recognizable. Now that sounds a bit harsh, but that was just the way it was, it’s a fixed core time […]. And then I realized that we can’t work like this in the future. We must also develop other working time models.” (21): ”[In the old times] The customer came, was handed over [the product], and we said ‘Thank you very much, it was nice that you were here.’ And only at the customer’s explicit request would one of our employees go outside and install it. So I said we can’t stand still there, we’re a service company and today we simply have to offer more. Then we started with service technicians.” (2): “So we found that the main problem […] was communication between the levels. The company was classically structured […]. Coordination […] took place vertically, but not horizontally. As a result, there was a lot of friction. […] We have solved this problem, […] reorganized the groups.” (6): “We experienced some dire straits, …because we were clinging to formerly successful structures whose success was individual-related.”
Theme 2: Intention and justification of refit actions	(12): “I just think that we decided what was best for the company at that time and then implemented those changes.” (23): “I already had an insight into the numbers and they were very bad at the time. […] My point now was to say what can be done so that we can get feasible business figures.” (1): ”[On change of staff] Then there are also people who don’t want to learn and they should have nothing to do here. […] I […] cut out dead wood.” (4): “I introduced a Monday management meeting […] for the management circle […]. It was pretty unstructured before. […] But for organization, it was very important to structure the employees. […] So I said, we’ll do that – if they only save 0.1% then they’ve already rationalized themselves […]. And in the course of this, we restructured […] the entire purchasing process.” (10): “We have changed the management structure. […] In the past, it was […] very status-related […] And we gave up this status thinking and said, that can’t be the motor and the purpose of such a body, it has to be functional.” (17): “I immediately changed the IT – that means we had almost zero IT equipment at the time. The only thing we had was a fax and a telex machine, and I then started to introduce a system, simply for daily work processes, from purchasing, to production and material procurement, to shipping and freight documents. In order to improve and also optimize these work processes.”
Theme 3: Performance effect of refit actions	(14): “A lot had settled around it, we had 75 employees, but we figured out that we could do the same job with 50.” (18): “The whole structure was in a sort of daily grind, […] now that we were initiating changes and new approaches, […] it was like a wake-up call which went through the enterprise.” (4): “The goods received were booked by the incoming goods department by hand, everyday hundreds of book entries. […] We now have completely new logistic systems. That had an immense impact.” (7): “We have set up an ancillary IT system that did not previously exist in this form […]. If I compare the amount of effort that was put into writing offers and invoices in the past, we have a very clear improvement […] that we only need 0.8 of an office staff instead of 1.5.” (31): “Before I took over, we had the following idea in the company – the bookkeeping should be 100% accurate. […] We had more people in administration and accounting than in sales. This had the following advantage – we always had perfect bookkeeping, which was never objected in tax audits because there were simply no errors. But that was over-administration, which doesn’t pay off for a wholesaler these days. I simply had to step in there, too. When I started there were over 20 people in accounting and administration, now there are 2.5.”
Theme 4: Successor’s ability (human capital) to refit	(10): “You are always learning something new and have to adapt and adjust to new situations. And then reflect for yourself whether what I do or the way I am fits my environment or not.” (2): “I wrote my diploma thesis about the company, about the existing structure, and the possibilities for improvement. And based on the things that I then looked at academically, we implemented various changes.” (8): “So I actually derived the whole topic from my studies […] You have to analyze and optimize the process.” (13): “[Reflecting on how experience helped upon succession] I worked first with corporations, and then with some medium-sized companies. After that, I took over this company. […] I completely rebuilt the thing!” (17): “If you come and already have a certain education and experience and can make certain decisions and people see that has quality, that’s a completely different weight.”
Theme 5: Turnaround (high need for refit)	(24): “All the numbers were red when I came in, we had to make a turnaround. […] There were some ideas, but no strategy per se. To tell the truth, everything I have done was thanks to the things I learned on other jobs prior to succession. […] Since the succession we have only made changes, so for eight years it has been a process of change for the whole company, all kinds of change.” (26): “We basically started to change a lot of things because it was ‘5 minutes to midnight.’ So, for the first time ever in the history of the company we had to lay off people. So basically, we took the firm, we turned it upside down, shook it up, and put it back to see what was left, which means I really started a highly dynamic change process … ” (18): „Because I simply believe in the company and was able to recognize what potential we have, especially through the crisis.”

(14): We used to have three tax and accounting consultancies, one for our assets and the wage payments, the next for the taxes and the third for public accounting. And still we factorized our bills while the payroll accounting was calculated by hand. […] The internal organization was really 30 to 40 years old, and it was about time to overhaul it. […] There was no way to communicate with the suppliers by e-mail or to communicate at all. The project leaders couldn’t write any letters themselves, so they dictated their bills on the dictaphone and then gave them to the typist. That’s not even 5 years ago. And the most crucial employee here at the company, was a lady who was able to write letters faster than the rest of the workforce. It might be funny, but it was actually the case. And the bottleneck was always when the lady was on vacation, no bills could be written and then there was no payment for 4 weeks and the liquidity was correspondingly bad during this time.

(18): We introduced a quality assurance department. It wasn’t there before, although it’s very important. […]. That doesn’t mean that he [the predecessor] didn’t deliver quality, but just that you can double check and times have changed. If you’ve been in a company for a long time, you might become a little blind to the company.

An abundance of instances in the qualitative interview suggests that the intention and justification of interviewed successors for their implemented changes indeed had an efficiency-oriented refit character (see Table 6, Theme 2)⁶

(12): I just think that we decided what was best for the company at that time and then implemented those changes.

Also, in several instances a positive performance effect and salvaged improvement potential were attributed to the post-succession refit actions that the CEO successors carried out (see Table 6, Theme 3):

(14): A lot had settled around it, we had 75 employees, but we figured out that we could do the same job with 50.

(18): The whole structure was in a sort of daily grind, […] now that we were initiating changes and new approaches, […] it was like a wake-up call which went through the enterprise.

The qualitative interviews also confirm that various elements of CEO-related human capital enabled successors to realize and reflect on (10) as well as leverage (2) improvement potentials (see Table 6, Theme 4):

(10): You are always learning something new and have to adapt and adjust to new situations. And then reflect for yourself whether what I do or the way I am fits my environment or not.

(2): I wrote my diploma thesis about the company, about the existing structure, and the possibilities for improvement. And based on the things that I then looked at academically, we implemented various changes.

Moreover, the information from the collateral open interviews also yields evidence that the above coherences are particularly present in a context where there is a high need for refit activity, such as when a succession takes place under turnaround conditions. For instance, CEO successors note (see Table 6, Theme 5):

(24): All the numbers were red when I came in, we had to make a turnaround. […] There were some ideas, but no strategy per se. To tell the truth, everything I have done was thanks to the things I learned on other jobs prior to succession. […] Since the succession we have only made changes, so for eight years it has been a process of change for the whole company, all kinds of change.

Overall, our main qualitative results echo and shed further light on the quantitative ones and are in line with the arguments put forward in our hypotheses (H1–H4).

Ancillary qualitative insights

Beyond this study’s focus on implemented refit actions, there may be several reasons why successors report a lower sum of implemented refit actions. Arguably these include (A) barriers to identification of necessary refit actions, (B) reasons for hesitating to implement refit actions, and (C) reasons for the failure of attempted refit actions. Accordingly, our qualitative interviews provide insightful ancillary evidence that there may be barriers surrounding multiple domains that potentially infringe the CEO’s recognition of a potential need for refit in the first place (see Appendix, Table A1, Theme A).

(13): But the problem with such a succession is that you are observing from the outside and therefore you don’t notice many things. That’s very clear. And then of course you are under a certain pressure when you are inside, you have to get things going and secure your existence.

Also, in some instances, we find evidence that successors did not attempt (or were hesitant) to implement refit actions despite recognizing a need for refit (see Appendix, Table A1, Theme B):

(28): Of course, there are also stories or company myths that influence the whole picture. Above all, if you know how the family feels about it […] So, if you always hear, for example, that the employees are part of the family and that is why employees have hardly ever been laid off and you are now in the situation where you say, well, seen from a purely rational point of view, people should now be laid off, then of course that always resonates in the back of your mind and accordingly it can also be a hindrance.

(7): No, well, my first goal was to create a smooth transition. That’s why I didn’t introduce new products at the beginning.

Finally, some successors noted they failed when trying to implement refit actions (see Appendix, Table A1, Theme C):

(21): We worked very closely with them [the new supplier] for a year and then it was canceled again. Actually […] for reasons that have to do with [our own] employees.

Together, these offer a deeper view beneath the surface of implemented measures, especially on the complexity during succession in family firms, and thus the ability required for their mastery.

Discussion

Why are successions so important to family firms? Certainly, one major aspect is the challenge of sustaining family leadership across generations (Ahrens et al., 2019). Indeed, continuity at the top of the firm is an important element of strategy for the long-run because the long-term approach also applies to relationships (Miller & Le Breton-miller, 2006). Such a strategy is known to yield a competitive advantage which is sourced by a cohesive firm culture and a garnered richness of social capital (Arregle et al., 2007; Le Breton-Miller & Miller, 2015). Consequently, family firm succession literature has focused on leadership and the family. Its most frequent topics are: incumbents and successors, the transfer of leadership between them, successor nurturing and important successor attributes, succession planning and family dynamics, timing and communication, and governance during succession (Le Breton–Miller et al., 2004).

This article contributes an additional, different perspective, a succession perspective that focusses on the organization’s realignment. This perspective is often overlooked, less discussed, and counterintuitive because the natural emphasis is on continuity, stability and long-term orientation in family firms as part of their success strategy (Miller et al., 2008). Our study shows that succession particularly matters for family firms because it may result in a period of fruitful change, rejuvenation and adaptation. This organizational change in turn may lay important foundations for the next long-run of the generation that has just succeeded into office. Our research provides evidence that able family firm successors often leave no stone unturned in the post-succession period. They introduce an astonishing number of reconfigurations across all levels and functions of the firm. Moreover, our analysis shows that this behavior significantly drives post-succession performance, and especially when the firm has a high need to adapt. These alterations are in the overwhelming majority of cases not a ritual of symbolic meaning that entails no performance effect (Gamson & Scotch, 1964; Meindl & Ehrlich, 1987; Pfeffer, 1981), rather the supported logarithmic form of the refit-action-performance relationship hints at decreasing marginal benefits as improvement potentials become increasingly salvaged by CEO successors through refit actions and performance restores (Donaldson, 1987; van de Ven et al., 2013).

It deserves mentioning that it is particularly difficult and complex to introduce and lead these adaptations in the family firm context – a context that treasures mindfulness regarding stability, social capital, and coherence, in particular with regard to managing the business for the long run (Miller & Le Breton-Miller, 2005) because the business has sustained, seasoned, and commitment-laden relationships with employees, suppliers, and customers. It may entail major discomfort and labor because adapting often involves not pursuing or differently pursuing what previous generations (often parents) have established (Harris et al., 1994). Nevertheless, our evidence suggests that many family firm successors are surprisingly active in these refit processes, and seem to accept and master the “Challenge of Entering the Green Room” (Schein, 1993).

Indeed, this work’s theoretical view on successions in family firms that rest on the conjuncture of CT (Donaldson, 1987; Le Breton-Miller & Miller, 2016) and UET (Ahrens et al., 2018; Cannella et al., 2008) contributes a more nuanced understanding of why succession for family firms is so important (Nordqvist et al., 2013) and why family firm successors are indeed often left with few other choices than to become such “Heroes of the Green Room” as previously unnoticed improvement potentials have silently built up over the time when a preceding generation was in office. The necessary, almost “natural inertia” that accompanies a management for the long-term approach because of its emphasis on continuity and stability that includes long reigns as a CEO but also its pronounced firm culture (Chirico & Nordqvist, 2010), makes succession a special window of opportunity for family firms. Accordingly, a refit through CEO turnover at the firm’s helm is possible and a previously accumulated organization-environment misfit can be dissolved via refit actions, provided this potential is recognized and adaptation is initiated by able successors (Cannella et al., 2008; van de Ven et al., 2013). This process arguably occurs under the boundary conditions that caring successors try to adhere to the obligations connected with preserving the firm’s vital social capital and reputation (Adler & Kwon, 2002; Ahrens et al., 2019). It then sets the stage for a new period of continuity that potentially covers the career of this next generation.

Our CT-UET inspired view, which sees family firm succession mainly as an instance of intense if not radical refit, may also provide deeper explanations for the puzzling empirical finding that family firms do not become stagnant, despite their emphasis on continuity, community, and connectedness via a long-term approach to management (Miller et al., 2008). Moreover, this theoretical conjuncture underlines why “family firms do not have to be characterized as inflexible, resistant to change, and burdened by traditions” (Hall et al., 2001, p. 206). Indeed, the renaissance processes that this CT-UET view postulates for family firm successions may well be a central element in the longevity of family firms (Zellweger et al., 2012).

Finally, at the CEO-level, this CT-UET view adds another crucial reason for choosing an able and apt family firm successor who is well equipped with CEO-related human capital (Ahrens et al., 2019; Calabrò et al., 2018): the ability to recognize organization-environment misfit and to implement refit in the critical juncture that the post-succession period in family firms embodies (Donaldson, 1987; Hambrick & Mason, 1984). In this respect, our findings inform scholarship on CEO attributes (in particular ability) and post-succession strategizing by the CEO, and on its downstream performance implications (Cannella et al., 2008; Hambrick & Mason, 1984).

In particular, while there is broader scholarly consensus that CEO-related human capital constitutes a mission critical predictor of firm performance – and especially that during CEO succession a deficit in CEO-related human capital materializes in substandard firm performance (Adner & Helfat, 2003; Ahrens et al., 2019; Bennedsen et al., 2007; Cannella & Rowe, 1995; Daspit et al., 2016; Holcomb et al., 2009; Le Breton–Miller et al., 2004; Pérez-González, 2006), this research contributes – by conceptualizing post-succession refit actions as a mediator – a deeper understanding of a pivotal mechanism behind this coherence. Indeed, the quantitative and qualitative evidence corroborating H1–H4 (summarized in Figure 4) suggests that CEO-related human capital positively affects the CEO’s capacity to recognize and address configurational shortcomings limiting the organization’s performance through refit actions.⁷

Figure 4. Summary of the findings.

H4 denotes the moderated mediation hypothesis. Significance levels: ***p < .01, **p < .05, *p < .1  

Thus, beyond improving the smoothness of the transition (Daspit et al., 2016; Morris et al., 1997), we detail a further layer of importance of the CEO’s attributes during family firm succession (Chrisman et al., 1998). This is, namely, that human capital-related factors eminently affect the CEO’s role as the “chief cognizer and decision maker” (Nadkarni & Herrmann, 2010, 1050) in the post-succession period. Our study of family firm successors suggests that when a CEO’s cognitive base is augmented by higher levels of CEO-related human capital, it allows her to (behaviorally and interpretatively) conceptualize a more rigorous mental representation of the organization and its fit with its environment as well as to more comprehensively recognize viable ways of improving fit via refit decisions (Cyert & March, 1963; Hambrick, 2007; Holcomb et al., 2009; March & Simon, 1958; Simon, 1955; van de Ven et al., 2013). This evidence not only provides detail to general strands focusing on managerial ability (Adner & Helfat, 2003; Cannella & Rowe, 1995), but also informs family firm specific strands on succession and inertia (Chirico & Nordqvist, 2010; Erdogan et al., 2020; Radu-Lefebvre, 2021).

Specifically, although “successors of multigenerational family firms are lifelong prisoners of a persistent moral dilemma: how to preserve and honor the company’s history and the business family’s past without jeopardizing its future” (Radu-Lefebvre, 2021, p. 173), our evidence advances that CEO-related human capital might allow the CEO to master this dilemma by introducing meaningful change in an inherently inert setting (Chirico & Nordqvist, 2010). Thereby, we provide explanations of why “a traditional organization is not necessarily opposed to meaningful change” (Erdogan et al., 2020, p. 46): for example, when able successors acting as behavioral⁸ agents of change revitalize and reinvent the firm during critical junctures, such as the window of opportunity succession provides.

Lastly, on a broader level, our proposed theoretical view, resting on contingency and upper echelon theories, entails wider relevance beyond the above portrayed CEO-level and post-succession period firm-level implications with regard to change. In our economies around the world, the most frequently occurring form of firm organization is the family firm (La Porta et al., 1999). Successions in family firms are turbulent, but inescapable occasions given the nature of humans as aging and mortal beings, and indeed only 30% of family firms survive past the lifespan of their founders (Ahrens, 2020; Sonnenfeld & Spence, 1989). Thus, even if the ubiquitous problem of finding willing and motivated successors is solved (Parker, 2016), poorly managed successions are known to plague the survival prospects of family firms (Miller et al., 2003; Morris et al., 1997). It is our hope that this theoretical lens, which is especially apt for visualizing the interplay of behavioral CEOs with the organizational dynamics during family firm succession, might provide an additional theoretical angle for future research that ultimately aims to foster our economies by making the human organization (a little) more robust to succession.

Our study does not come without its limitations. It deserves mentioning that our quantitative data covers implemented refit actions. It does not capture the successor’s intentions to implement refit actions nor failed attempts to implement refit. Thus, fruitful avenues for future research include (a) to analyze which successors are failing to recognize a need for refit actions in the first place (for instance via experimental studies), (b) to shed more light on why successors who recognized a need for refit did not attempt to address them, (c) to identify why refit initiatives by CEO successors failed during their attempt to implement them. In this respect, our ancillary qualitative evidence already suggests that indeed all three (a, b, and c) affect the post-succession period in important ways. These side findings deserve further quantitative scholarly scrutiny, especially with respect to the “black box” of underlying cognitive and affective processes that lead to the successful perception of misfit and execution of refit. Relative to this, now that we have documented that CEO-related human capital increases implemented refit actions, future research should inquire whether CEO-related human capital equally enhances underlying perception (a) and attempt and success rates (b and c). Similarly, this work has documented that across the cases we observe not all refit actions are equally important to performance, and while their sum entails a log-positively relation to performance, few have a significant performance effect alone. This suggests two coherences: an implementing successor seems to prioritize those actions that seem to entail the most refit potential, while refit seems to increase especially when several changes are implemented in conjunction. While beyond our scope, there is merit in future research which delves deeper into these patterns. Finally, our study is based on German family firm succession cases, thus a typical “Western” context. Hence, to achieve more generalizability, more scholarly scrutiny beyond this institutional context is needed: Future research might entail cases from different contexts which vary in their emphasis of tradition, culture, and stability, in other words contexts that vary in the social adequacy of change. Indeed, although our research finds that post-succession changes are not often rituals of symbolic meaning without any performance effect (Gamson & Scotch, 1964; Meindl & Ehrlich, 1987; Pfeffer, 1981), future research might indeed inquire into a potentially ritualistic role of change during family firm succession in other cultures, as the respective patterns we observe in this regard might be germane to the cultural context of our study.

Practical implications

We emphasize that our findings should not be confused with a “blueprint rule” to introduce change upon family firm succession. Instead, we highlight the crucial role of a thorough understanding of an individual firm’s situation before initiating change, and CEO-related human capital is decisive for this. When following the preceding generation’s footsteps, the new CEO not only inherits a formal position, but also finds a current internal and external equilibrium of organizational rules, norms, power structures and procedures, which are, in the abstract, nothing other than the result of the organization’s past experiences and behavior. The question is this: are the reasons for the current organizational configuration still extant, and if not, are there better configurations achievable within the limits of reasonable effort? The new CEO often has to combine the best of both old and new worlds, the organization’s past experience with (possibly external) insights rooted in the present reality to unleash the additional performance potential of her company and to restore its organization-environment fit.

Conclusion

Inspired by contingency and upper echelon views, our study shows that family firm successors can achieve higher performance upon succession by utilizing refit actions to address accumulated improvement potential arising from a sub-optimal organization-environment fit prior to succession. Our results highlighting the CEO related human capital as an antecedent of such refit actions also emphasize the vital role this important successor characteristic plays in family firm succession context. As our results reveal, the vital role of human capital is especially evident when the succession is characterized by a turnaround situation, increasing the complexity and uncertainty associated with post-succession processes. Although not free of limitations, our study offers important contributions to the theory- and practice-oriented discussions revolving around family business successions and post-succession performance in family firms.

Acknowledgments

We would like to thank the handling editor Esra Memili and anonymous reviewers for their valuable comments and improvement suggestions. We would also like to thank Kyung eun (Trixie) Park and Marc Kowalzick for their friendly input to the prior versions of this article. We are also thankful for the feedback we received on the prior versions of the paper from several international conferences including 74th Academy of Management Annual Meeting (2014, Philadelphia, U.S.).

This article holds the “BDO Best Contribution to Practice Award” from 5th International Family Business Research Forum (2022, Hasselt, Belgium) as “Fusing Contingency Theory & Upper Echelon Theory: For a New Perspective on the Critical Juncture of Family Firm Succession.” A prior version of the article was also awarded with “FFI Best Unpublished Research Award 2014” by Family Firm Institute (FFI) - Global Conference (2014, Washington, US) under the title: “Heroes of the Green Room - Post-Succession Restructuring and Corporate Performance in Family Firms.” It was also nominated for a best article award by the 22nd Interdisciplinary Annual Conference on Entrepreneurship, Innovation, and Mittelstand (2018, Stuttgart, Germany) as “The Role of Change in CEO Succession in Family Firms.”

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ To detail this metaphor, Schein (1993, p. 87) writes: “If you put a dog in a green room, ring a bell, and ten seconds later give the dog a painful electric shock in that room, the dog will fairly quickly learn to avoid green rooms, and, if he hears a bell, will try to run away or if he is restrained, will cower anxiously. Even if you turn the shocks in the green room off, the dog will not enter it and, therefore, will not discover that the shock is off.”

² In that sense, family firm succession translates to a setting where it became “necessary for the dog to learn to live in a green room full of bells” (Schein, 1993, p. 87).

³ Certainly, many other scenarios upon a succession are possible. These would also include various scenarios where successors would (be able to) identify and address only a portion of the accumulated improvement potentials. For the sake of argument and simplicity, we focus on two distinct scenarios representing the two extremes where these improvement potentials are either identified and addressed, or not.

⁴ In order to calculate both adjustments, we harness an adjustment group retrieved from the Amadeus database that consists of more than 187,000 company-year observations. The industry-adjusted values are calculated by a subtraction of the median PM of the respective year and industry (two digit ISIC level) of the Amadeus database firms from our sample firm values. In line with Pérez-González (2006), performance adjustments are conducted by dividing the industry-adjusted PM of the adjustment group firms into performance deciles for each year. We then matched the industry-adjusted PM of each sample firm with the respective performance decile in the year of the succession to find the relevant performance comparison group for each sample firm. In a next step, we subtract from the industry-adjusted PM of the sample firms the median industry-adjusted PM of the relevant decile and year (trend/decile development). Unadjusted PM values are winsorized at the 0.025 level.

⁵ This implies that across cases the individual refit actions are likely to have different performance implications depending on the local case-specific scenario, a circumstance which is mitigated by the summation approach we took. Also, the above variable is geared to measure the sum of implemented refit actions. Although it seems plausible to assume that the higher the sum of refit actions the higher the CEO’s perception of the need for fit, failed or un-attempted implementations might additionally explain a lower sum (we follow this line of thought in our ancillary qualitative analysis as well as in Appendix A1). Also, by testing for the positive relation of the sum of refit actions with post-succession performance, we implicitly test that these actions are intended to increase the fit, thus thereby largely reducing the role of alternative explanations, such as symbolic action (Pfeffer, 1981) which we further corroborate in ancillary qualitative interviews. Moreover, in robustness tests, we also harness a narrower measure.

⁶ In robustness checks, we screened the ancillary qualitative interviews for passages that could potentially also hint at ritualistic action and only found one successor (12) who noted on power and changes: “But ultimately it was also very important to make it very clear […] where I stand in the organization of the company.” Otherwise, we cannot report any support in this respect.:

⁷ Moreover, by investigating how this mediation is moderated by the factual existence of a misfit, we “reveal additional contingencies surrounding the effects of managerial ability” (Holcomb et al., 2009, p. 480).

⁸ In the sense that they act as constrained interpreters (UET) of the everchanging contingency facing the firm (CT).

Appendix

Table A1. Exemplary quotes for ancillary qualitative insights.

Theme	Exemplary Quotes
Theme A: Barriers to identification of necessary refit actions	(13): “But the problem with such a succession is that you are observing from the outside and therefore you don’t notice many things. That’s very clear. And then of course you are under a certain pressure when you are inside, you have to get things going and secure your existence.” (30): “In our industry, decisions are very long-term. […] Which vineyard, which grape variety do we plant? These are decisions you always make for at least a generation and you don’t know if they’re right or wrong. […] It is like it is.” (6): “Oh, I wasn’t aware of it [that there would be such a number of change initiatives required], you maybe suppress such thoughts at the beginning. When everything is going well, you don’t think about it.” (24): “If you’re in a cold [low discretion] position where you are not calling the shots, then you start believing [the story] and probably do exactly the same thing.”
Theme B: Reasons for hesitating to implement refit actions	(28): “Of course, there are also stories or company myths that influence the whole picture. Above all, if you know how the family feels about it […] So, if you always hear, for example, that the employees are part of the family and that is why employees have hardly ever been laid off and you are now in the situation where you say, well, seen from a purely rational point of view, people should now be laid off, then of course that always resonates in the back of your mind and accordingly it can also be a hindrance.” (7): “No, well, my first goal was to create a smooth transition. That’s why I didn’t introduce new products at the beginning.” (8): “And at the end of the day you have to wait until one or the other retires to get the issue resolved. Taking something away from someone who has done it for 30 years is very complicated.” (6): “I mean, sending the 10 people away, I was concerned about that for weeks, before the decision, after it and […] I went through the suffering which of course is not comparable to what the people experienced when they are told about it.” (15): “Actually, if I’m completely honest, I should have put his resignation on the table 5 years ago. Why didn’t I actually do it? It was uncomfortable and so I organized around it.” (8): “If we introduce that now, then you have to assume that two-thirds of your middle management will leave.” (19): “But the law stipulates that it’s [laying off] not that easy either. So you just can’t act [implement the change] the way you want to.” (19): “But the things that have to do with financial expenses are very, very difficult. […] How many are working down there? That is 30 people and only one toilet. That’s a point. […] But all that comes at a cost and that’s why you can’t change it quickly.” (22): “Because as long as the predecessor is in the company, you don’t really have a chance as the successor. […] You can’t really change anything or intervene, because all of your changes will have an implicit connotation of criticizing the predecessor’s work.” (31): “They [employees and certain executives] are not going along with these changes. […] Here, too, you have to wait for certain personnel changes [due to retirement] in the company, because otherwise it would be too difficult or too exhausting.”
Theme C: Reasons for failure of attempted refit actions	(21): “We worked very closely with them [the new supplier] for a year and then it was canceled again. Actually […] for reasons that have to do with [our own] employees.” (22): “You can do whatever you want – if the employees don’t stand behind you, then you can buy as many machines as you want, or spend as much money as you can – it won’t work. They will drive your machines to the wall, they will ruin your production – it just won’t work.” (22): “Certainly I have to live with the accusation that some things [changes] went wrong, and why didn’t I take care of them more intensively. […] But, at some point you reach your personal limits.”