«J Bus Econ (2015) 85:793–815 DOI 10.1007/s11573-014-0753-1 ORIGINAL PAPER New insights on CEO charisma attribution in companies of different sizes ...»
J Bus Econ (2015) 85:793–815
New insights on CEO charisma attribution
in companies of different sizes and ownership structure:
the role of prior company performance
Bjorn Michaelis • Florian Kunze • Heike Bruch
Published online: 23 October 2014
Ó Springer-Verlag Berlin Heidelberg 2014
We extend theories on charismatic leadership by investigating the
inﬂuence of prior company performance on subordinates’ attributions of chief executive ofﬁcer (CEO) charisma within companies of different sizes and ownership structure. First, we use an experimental design to examine the effects of prior company performance on attributions of CEO charisma. Second, in a ﬁeld study with 69 companies we replicate the experimental ﬁnding and show that this relationship is moderated by the size of the company such that the relationship between prior company performance and attributions of CEO charisma is signiﬁcant only in large companies. We ﬁnd no evidence, however, that the ownership structure of a company could strengthen or weaken this relationship.
Keywords CEO charisma attribution Á Company performance Á Company size Á Ownership structure M1 JEL Classiﬁcation B. Michaelis (&) UBS-Endowed Professorship for Strategic Management, Goethe University Frankfurt, ¨ Gruneburgplatz 1, 60323 Frankfurt am Main, Germany e-mail: email@example.com F. Kunze ¨ University of Konstanz, Chair for Organisational Studies, Universitatsstraße 10, 78464 Konstanz, Germany e-mail: ﬂorian.firstname.lastname@example.org H. Bruch University of St. Gallen, Dufourstrasse 40a, 9000 St. Gallen, Switzerland e-mail: email@example.com 794 B. Michaelis et al.
1 Introduction Over the last two decades, charismatic leadership theories highly inﬂuenced research on organizational leadership (Conger and Kanungo 1987, 1988a, b; Shamir et al. 1993). In particular, Conger and Kanungo’s (1987) charismatic leadership paradigm was highly inﬂuential (Galvin et al. 2010) by explaining that subordinates’ perceptions of their leaders’ behaviors determine whether they see their leader as ¨ having charisma (e.g., Karreman et al. 2006; Nohe et al. 2013). Numerous studies have shown that charismatic leaders are more effective than their non-charismatic counterparts (e.g., Atwater et al. 1991; Howell and Frost 1989; Koene et al. 2002) and that charismatic leadership may lead to higher subjective and objective performance outcomes on different levels of analyses (e.g., Bono and Ilies 2006;
Lingenfelder and Wieseke 2010; Michaelis et al. 2009; Waldman et al. 2001).
On the organizational level, substantive research has investigated how chief executive ofﬁcer (CEO) charismatic leadership attribution is related to company performance (e.g., Tosi et al. 2004; Waldman et al. 2004). However, most of that
research neglected to consider the possibility of a reversed causal relationship:
whether a company’s prior performance may engender higher levels of CEO charisma attribution (for exceptions see Agle et al. 2006; Awamleh and Gardner 1999). For instance, neither Waldman et al. (2004) nor Tosi et al. (2004) controlled for prior company performance, although various scholars (e.g., Virany et al. 1992;
Waldman et al. 2001) bemoaned the lack of such controls. Waldman et al. (2004) drew their conclusions on associations between charisma and prior company performance merely on correlations without controlling for variables such as company size (Agle et al. 2006). This is surprising because company size determines executives’ decision-making discretion (Hambrick and Finkelstein 1987), so that prior company performance might have different impacts on CEO charisma attribution in ﬁrms of different sizes.
Empirical evidence on the relationship between prior company performance and CEO charisma attribution is rare, incomplete, and so far based on only two studies.
One study found a positive relationship between organizational performance and CEO charisma attribution (Agle et al. 2006), although they used a non-experimental design to test the assumptions and thus avoided the question of causality (Shadish et al. 2002). A second study also demonstrated that organizational performance positively affects CEO charisma attribution (Awamleh and Gardner 1999).
However, the authors emphatically emphasized that their results were based on an artiﬁcial design because only undergraduate students participated in the laboratory study. Furthermore, all studies relied on data collected in only one cultural setting, the United States, and most ignored contextual factors that might affect the proposed relation. In particular, prior studies used data from large publicly listed companies and thus could not answer whether company size (i.e., small and medium vs. large) or ownership structure (publicly listed vs. family-owned) are important contingent factors.
In consequence, our study strives to answer those open questions in the ﬁeld of charismatic leadership research by executing a constructive replication and extension of prior work (Eden 2002) in three ways. First, we further investigate
New insights on CEO charisma attribution 795
and clarify the relationship between prior company performance and CEO charisma attribution through a mixed-method design. For that purpose we ﬁrst applied a laboratory study and then cross-validated the ﬁndings in a ﬁeld sample of German companies of different sizes and ownership structure. Thus, to our knowledge, ours is the ﬁrst study exploring prior company performance as an antecedent of CEO charisma attribution with an experimental design combined with a ﬁeld study in a European sample. Consequently, in contrast with prior studies, our study design allows conclusions about causal effects and important implications for real business settings.
Second, acknowledging criticisms of prior research for failing to include ﬁrm size as a focal variable (Agle et al. 2006), we investigated two potential organizational-level boundary conditions for the impact of prior company performance on CEO charisma attribution. Speciﬁcally, our ﬁeld study tested whether company size (i.e., number of employees) and ownership structure (i.e., shares held by family members) can strengthen or weaken the impact of prior company performance on CEO charisma attribution. Thereby, our study offers signiﬁcant theoretical and practical implications for scholars and executives on how to transfer and apply charismatic leadership concepts to companies of different sizes and ownership structure.
Finally, establishing prior company performance as a prominent antecedent of charisma attribution extends the theoretical framework of charismatic leadership (Conger and Kanungo 1987), which neglected performance and focused rather on leader behavior as the source of charisma attribution. By considering prior company performance, we agree with suggestions (Agle et al. 2006) to further test the dimensions of the theoretical framework.
2 Theory and hypotheses
2.1 Charismatic leadership The word charisma comes from charismata, ancient Greek for ‘‘gift’’ (Conger and Kanungo 1992). Sociologist Max Weber [(1924) 1947] was the ﬁrst to apply the word to the scientiﬁc/secular leadership context. Conger and Kanungo (1987) modiﬁed Weber’s conception and deﬁned charisma as ‘‘a relationship between an individual (leader) and one or more followers based on leader behaviors combined with favorable attributions on the part of the followers’’ (Waldman et al. 2001: 135).
We seek to validate and further develop Conger and Kanungo’s (1987) charismatic leadership model. In testing their assumption, we aim to research whether prior company performance is a key prerequisite determining whether subordinates will attribute their CEO with having charisma. Furthermore, we test whether certain organizational-level boundary conditions (i.e., company size and ownership structure) affect this relationship. The theoretical rationales for the relationships are developed in the following two sections.
2.2 Prior company performance and subsequent attributions of CEO charisma Many studies have examined charismatic leadership effects on subjective and objective performance outcomes at the individual, team, and organizational levels (e.g., Michaelis et al. 2009; Rowold and Laukamp 2009; Waldman et al. 2001). For instance, CEO charisma was shown to have possible long-term effects on company performance through its effects on internal performance and external relationships (Lowe et al. 1996). However, those studies mainly considered the effects of charisma on performance and neglected to investigate the effects of prior performance on subordinates’ attributions of charisma. For instance, one study suggested that subordinates’ knowledge about leaders’ prior performance can inﬂuence charisma attributions (Puffer 1990), which raises the possibility of reverse causation: leaders’ performance as the cause of charisma attributions (Kirkpatrick and Locke 1996). However, only a few studies have examined the inverted direction of causation (e.g., Awamleh and Gardner 1999), with mixed empirical evidence and no ﬁnal conclusions (Agle et al. 2006).
In our study we build on the observation that individuals attempt to understand their environment by attributing events and outcomes to prior causes. Organizations are complex environments that include ambiguous and multi-determined events that challenge individuals to make reasonable and understandable causal attributions.
Subordinates searching for causes for organizational events are likely to look to the quality of their leadership for making causal attributions (Shamir 1992) and for attributing complex organizational outcomes. Organizational outcomes include, for example, prior organizational success or failure (Meindl et al. 1985). Consequently, this ‘‘attributional sense-making’’ process is increased by ‘‘fundamental attribution error’’ as individuals tend to overestimate personal factors such as leader behavior and underestimate situational factors such as general economic situations. The general attribution model (Kelley 1967), attribution theory of leadership (Calder 1997), and empirical research (Meindl et al. 1985; Shamir 1992) provide evidence that perceived organizational performance leads subordinates to attribute success to the leader, particularly to the CEO at the top (Shamir 1992).
Those ﬁndings and assumptions ﬁt neatly into Conger and Kanungo’s (1987) attribution-based model. Considering how closely their model approximates Weber’s (1947) conceptualization, it is somewhat surprising that their theory has not yet integrated the performance element, although repeated success—as a validation of the leader’s exceptional character—is crucial for producing and maintaining perceptions of charisma (Weber 1947). Hence, performance could be an unidentiﬁed factor in Conger and Kanungo’s approach, one that also leads to charisma attributions. Thus, we assume that prior company performance is
positively linked to attribution of CEO charisma, which leads to our ﬁrst hypothesis:
H1: Prior company performance will be positively related to subordinates’ attribution of CEO charisma.
This relationship might differ, however, in companies of various sizes and ownership structures. For that purpose we next consider the contextual role of company size and ownership structure.
2.3 The contextual role of company size We argue that company size plays a crucial contextual role in the relationship between prior company performance and CEO charisma attribution. Company size signiﬁcantly alters employee interactions and perceptions about top managers (Hambrick and Finkelstein 1987). For instance, small companies typically derive their strategies informally, inexplicitly, and intuitively (Mintzberg 1988) mainly under CEO direction (Miller and Toulouse 1986). Moreover, CEOs in small ﬁrms are highly involved in decision-making processes and thus play a critical role in adopting new technologies (Lefebvre and Lefebvre 1992). Aligned with and supporting this reasoning, researchers have argued that in small companies with less-complex organizational contexts, senior executives may have more managerial discretion and consequently have more inﬂuence and direct interaction with their employees (Agle et al. 2006; Finkelstein and Boyd 1998; Hambrick and Finkelstein 1987); in larger, professionally managed ﬁrms, CEOs are more distant and interact indirectly with most employees, often only through middle management (Lubatkin et al. 2006); The different forms of interaction are likely to affect employee perceptions of the CEO, and will affect whether they see the CEO as charismatic.
In large companies most employees lack personal contact with CEOs and therefore cannot evaluate their charismatic qualities on interpersonal levels.
Consequently, they might take prior company performance as a substitute for attributing charisma to the CEO. In small companies, prior company performance might be less important for judging charisma, as most employees can use their direct personal interaction rather than substituting prior company performance.
In addition, environmental uncertainty in organizations and societies has major impact on charisma attributions (e.g., House et al. 1991). Smaller companies are clear and non-ambiguous: employees share the same expectations regarding appropriate response patterns and required knowledge and skills (Shamir and Howell 1999). In larger companies, employees feel more uncertain because they have less information and cues for judging appropriate behavior. More complex organizational structures and environments make subordinates more attentive to CEOs’ prior performance as a criterion for charisma attribution. Thus, subordinates in large companies will over-attribute prior company performance when perceiving
charismatic qualities. In sum, we propose a moderation hypothesis:
H2: The positive relationship between prior company performance and subsequent CEO charisma attributions will be stronger in larger companies.
2.4 The contextual role of a company’s ownership structure