In the eyes of the beholder : let's talk strategy and identity interpretation : a research on the discursive construction of executive personality factors and their influence on strategic choices

(1)

In the eyes of the beholder:

Let’s talk strategy and identity interpretation

A research on the discursive construction of executive personality factors and their influence on strategic choices

Master Thesis

MSc. Business Administration – International Management Supervisor: Dr. J.P. Lindeque

Second reader: Dr. M.K. Westermann-Behaylo Student: L.K. van Zeijst

Student ID: 5966930

(2)

1

Abstract

This thesis discusses how research on the Upper Echelons Theory and strategic discourse are related. The UET proposes a model of how an organization, its strategic behaviour and its performance are a reflection of its executives’ psychological characteristics as they directly affect the central process of strategic decision-making in an organization. The mutually constituting construction of the social reality in strategic discourse translated in three layers of analysis (i.e., a macro contextual level, meso practice level and micro textual level) shows resemblance to what is regarded as the problem of endogeneity and increasingly complicated causal relationships between the main constructs in the UET (i.e., Contextual environment, the strategic choices and the personality of the individual executive). The social process of decision-making and its interdiscursive relation with the perception of the personality factors attributed to the main executive is explored within a reconstruction of the strategic discourse. The outcomes of this thesis reflect how the construction of identities for the executive and their embedded personality factors affects the interpretation and expectation of strategic choices. This results in resistance, appreciation, valuation and support by members within the organization and outside analysts, shareholders and media.

Keywords: Upper Echelons Theory, Discourse Analysis, Decision making, HEXACO

(3)

2

Statement of Originality

This document is written by student Lenty Kyra van Zeijst, who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating

it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents

(4)

3

Acknowledgements

A master thesis is the grand finale of your studying life, before you get to put it all into practice. It is the crown on your work and a representation of your capabilities. It’s not the subject, nor the grade, what makes this project so relevant. It is the feeling you performed well all those years. It is the feeling that the nights you continued writing papers and the exams you skipped because you were enjoying a social life, were worth it the experience.

It feels like an end of an era. Since my heart problems emerged at the age of 13, there has only been one other consistent and continuing storyline through my life. Studying. It didn’t matter how sick I was, I always pursued challenging my capabilities. I started with a Bachelor in Communication science, and received the opportunity to study at Sciences Po in Paris and at USP in Saõ Paulo, Brazil. Now at the Business School in Amsterdam I hope to finish with my Master in Business Administration within the specialization International Management. This thesis was a confrontation with my own personality as I struggled with a strive for perfection and a burnout. Nevertheless, I am very proud of the result.

This thesis deserves a special acknowledgement for the input, support and patience of Johan Lindeque. Thank you for challenging me, cheering with me and keeping me sharp.

Further, I would also like to thank all those who supported me while I wasn’t participating in social life. Thank you, for your empowerment, inspiration, trust and patience. Thank you for being there knowingly or unaware. Without your words of encouragement, discussions on leadership and life, some of your actions of support and continuing presence, it would have been a lot harder. Thank you for the talks, the walks, the wining, the dining, the coffee, the cheering, the crying, laughing, the dying and the correcting.

(5)

4

Index of Tables and Figures

Table 1. Position overview ‘Tex Gunning’ ... 42

Table 2. Overview of Cases with Central Focal Point of Analysis ‘Tex Gunning’ ... 44

Table 3. Categorization of Texts ... 46

Table 4. Strategic themes ... 51

Table 5. Coding table category ‘Contextual environmental factors’ ... 51

Table 6. Coding table category ‘Strategic choices’ ... 51

Table 7. Coding table category ‘Trait HEXACO characteristics scale’... 52

Table 8. Coding Table Category ‘Author’ ... 54

Table 9. Coding Table Category ‘Decision-making elements’ ... 54

Appendix A. Timeline Texts ... 96

(8)

7 “By changing the definition of what is being studied, we change what we see; and when different definitions are used to chart the same territory, the results will differ, as do topographical, political, and demographic maps, each revealing one aspect of reality by virtue of disregarding all others”

(9)

8

1. Introduction

According to the Dutch newspaper Financieele Dagblad big data, the infinite source of information available due to digitization, is becoming more important for organizations concerned with ‘executive search’ (Schoemakers, 2016). Organizations increasingly are able to identify general patterns in attitude and behaviour, by correlating given facts with behavioural outcomes (Wang, Holmes, Oh, & Zhu, 2016). The utilization of big data is argued to enable recruiters to select more suitable candidates for a position, as these consistent structures of related variables accurately predict individual behaviour, norms and values, and competences (McAfee et al., 2012) and can avoid selection based on criteria limited to former work experience and individual assessment of personal behaviour (Wang et al., 2016). As an example the article describes how they assess risk aversive behaviour of executives based on, among other variables, the size and length of their mortgage (Schoemakers, 2016).

This method of ‘executive search’ is in line with the original theoretical framework and methodology of the Upper Echelons Theory (UET, Carpenter, Geletkanycz & Sanders, 2004) suggesting it enables organizations to select new executives in line with the preferred strategy of the organization, as executive demography, such as such as age, tenure and education, can be used as proxy for behaviour (Hambrick & Mason, 1984, Hambrick, 2007). The UET suggests that an organization, its strategic behaviour and its performance are a reflection of its executives’ psychological characteristics, as they directly affect the central process of strategic decision-making in an organization (Hambrick & Mason, 1984). The theory assumes that strategic choices result from rational decisions, which are an irreducible consequence of environmental stimuli (Cannella & Holcomb, 2005; Hambrick, 2007; Rost & Osterloh, 2010).

From the inception of the theory there were several theoretical and methodological challenges acknowledged (Wang et al., 2016). As executives’ had little time for

(10)

time-9 consuming self-evaluation methods to establish their psychological characteristics, observable demographic characteristics were suggested as proxies for these illusive concepts (Hambrick & Mason, 1984). Meanwhile a lack of empirical support for the real underlying causal processes related to the role executives’ characteristics in decision-making remained (Cannella & Pettigrew, 2001; Carpenter et al., 2004; Hambrick, 2007; König et al., 2014). Furthermore, as the body of research grew, antecedents, mediators and moderators were added to the theory, challenging the framework by stating that the correlation between constructs is often mistaken for causation due to the problem of reversed causality (Carpenter, Geletkanycz & Sanders, 2004), endogeneity (Hambrick, 2007) and lack of insight in the underlying psychological characteristics (Lawrence, 1997; McAfee et al., 2012). More comprehensive and dynamic approaches were recommended, to address these complexities and to uncover how the executive and organization are socially constructed in larger contexts (Carpenter, et al., 2005). These common limitations in UET literature are often referred to as the ‘black box’ of unknown, underlying psychological and social processes of individuals and organizations (Cannella & Pettigrew, 2001; Carpenter et al., 2004; Hambrick, 2007).

In response to these challenges of the UET, König et al. (2014) proposed to take on a new approach to 'open this black box' by analysing discourses. “Discourse analysis provides a theoretical and methodological framework for exploring the social production of organizational and interorganizational phenomena” such as strategic decision-making (Phillips, Sewell & Jaynes, 2008, p. 770). Research on strategy-as-practice (SAP; Balogun et al., 2014; Hardy & Thomas, 2014), which explores strategy as a situated socially accomplished activity, regards strategic discourses as the most prominent feature of strategy within the organization and highlights the importance of decision-making as the main element in a strategic discourse (Balogun et al., 2014; Hendry, 2000).

(11)

10 Additionally, discourse analysis is an increasingly popular method for examining how individuals constantly strive to shape their personal identities in organizations and are being shaped by discursive forces (Fenton & Langley, 2011; Sveningsson & Alvesson, 2003; Vaara, 2010). Contextual discourses surrounding the notion of strategy can be analysed to understand who is being constructed as a legitimate practitioner of strategy and what this might mean for organizational outcomes (Sveningsson & Alvesson, 2003). More specifically, how the upper echelons make sense of the strategic situation and give sense to information in the context of analyst evaluations is constructed into the representation and interpretation of the organizations’ strategic choices (Chanal & Tannery, 2011; Kaplan & Tripsas, 2008; Whittington, Yakis‐Douglas & Ahn, 2016).

Analysing this identity construction in the construction of strategic discourse as empirical material provides a significant yet unexplored opportunity for UET research (Brennan & Conroy, 2013; Vine et al., 2008; Walji, 2009). Beyond the use of unobtrusive measures for individual personality constructs, such as narcissism and internal locus of control (Hiller & Hambrick, 2005), little research is found within UET research that applies methods related to discourse analysis to understand how the psychological characteristics are constructed in the process of decision-making and within strategic discourse (Brennan & Conroy, 2013; Vine et al., 2008; Walji, 2009). Approaching the UET from a discursive perspective, this thesis adds focus on the non-instrumental phenomena that surround the executive and his influence on the strategic choices (Hambrick, 2007). By answering the following question this study intends to explore how and why executives' characteristics become manifested in organizational outcomes:

(12)

11 The research question is answered by a combination of qualitative research methods, such as qualitative multiple-case study research, unobtrusive measures for data collection, qualitative content analysis, to perform a multi-case discourse analysis. The central focal point within this thesis is the executive Tex Gunning, former CEO of Vedior, Managing Director and Boardmember at AkzoNobel, CEO of TNT Express and currently CEO of Leaseplan. This executive has a career and outspoken opinion on strategy and leadership that fits within the thesis objective. His appointment at the separate companies in similar positions creates a contextual pattern and allows performing a cross-case analysis between the discourses at the separate organizations (Eisenhardt & Graebner, 2007).

The literature review below explores the similarities between discourse analysis in organizational research and the UET. The two theoretical approaches are presented in an analytical framework, which consists of the three levels of discourse (text, discourse practice-context and macro-practice-context), the three main constructs in the UET and the central process of decision-making. The analytical categories relevant for text analysis emerged during an iterative process of going back and forth between the theoretical concepts developed in UET and discourse sections in the thesis and the retrieved empirical data described in the methodology section. A qualitative content analyses on the strategic and underlying discourses resulted in three cases in which the strategic choices are constructed and compared in a cross case analysis (Mayring, 2000; LeGreco & Tracy, 2009; Tracy, 2010).

(13)

12

2. Literature review

This section proceeds as follows, first are the two research approaches in this thesis, discourse analysis in organizational research and the Upper Echelons Theory, are briefly explained separately. Both the discourse analysis and the UET are described by the main elements in the theories are related, followed by a perspective on the main actors at play and the approach towards the executive as focal point in the decision-making process. Additionally research the main constructs of the UET will be converged with the discursive approach.

2.1. Discourses

Discourses are regarded as the general and prevalent systems for the formation and articulation of ideas in a particular period of time, constructing and maintaining a social reality (Alvesson & Karreman, 2000). The key assumption is that discourse is embedded in social reality and thus constitutes, constructs and reflects any phenomenon we are interested in by language, such as organizations, strategy and individuals (Alvesson & Karreman, 2011; Hardy et al., 2005). The construction of this social reality is seen as dialectical and mutually constituting, as discourse can be considered as both an object and a practice, simultaneously shaping the structures that constrain it (Phillips et al., 2008).

Discourse analysis offers a variety of approaches to capture the interrelationship between text and context and to unravel the different interacting processes shaping discourse (Phillips & Oswick, 2012). Discourse can be delineated on the basis of the focal point of analysis, such as identity, institutions, strategy or organizational change, and on the basis of level-based methodological approaches (Phillips & Oswick, 2012). Fairclough, a main contributor in research on discourses, has provided a framework incorporating three levels of discourse to deconstruct the process (1995; 2003) (See Model 1, Levels of discourse framework).

(14)

13 The model portrays a micro textual level to describe and interpret the discourse. On this level, the focus is on discourse as object, such as the specific linguistic features in discrete episodes or conversations, which can represent cognition and behaviour (Boje, Oswick & Ford, 2004; Pennebaker, Mehl & Niederhoffer, 2003). On the meso-level the interaction takes place. The production, distribution, reception and possible adaptation of texts are placed on this level as it represents the discursive practice (Beelitz & Merkl-Davies, 2012). It provides interpretation and explanation of the presence of specific discourses in the texts and how they are related to each other (Boje et al., 2004). The macro-level involves the wider perspective to explain how

Model 1. The levels of discourse framework

(15)

14 the discourse is constructed by situational, societal and institutional themes and trends. The social practices on this level contain grand narratives and mega-discourses with wider social implications (Fairclough, 1995; Ruiz, 2009). This macro level is used to explain the production and interpretation on the meso-level by the macro themes and trends. The processes on these levels take place simultaneously and in an iterative movement that resembles a continuous dialogue (Ruiz, 2009).

2.1.1. Strategic discourse

Strategy was one of the earliest areas of study in which organizational discourse was applied (Vaara & Whittington, 2012). The stream of research that combines the two disciplines of discourse analysis and organizational research is that of strategy-as-practice, which approaches “strategy as something individuals do, rather than as something the organization has” (Jarzabkowski, Balogun & Seidl, 2007, p.6). In this perspective strategic decisions are “identified as part of an organizational discourse or body of language-based communications that operates both at the structural and at the communicative levels, and that constitutes a central feature of the strategy process” (Hendry, 2000, p. 957). This approach of organizational ideas and practices as discursive and social construction had several implications (Fairhurst & Putnam, 2004; Vaara, Kleymann & Seristö, 2004; Vaara & Whittington, 2012).

It suggests analysis of the physical materialized resources resembling an organization’s strategy, such as transcripts, press releases and annual reports, which allows researchers to empirically assess the strategic processes (Balogun et al., 2014). Additionally it acknowledges that strategizing is a social activity, as communication requires different actors (Vaara & Whittington, 2012). Different perspectives shape discourses and executives with significant influence affect the subjectivity and power relations between actors engaged in strategizing (Hendry, 2000; Mantere & Vaara, 2008). The identities of executives are also

(16)

15 constructed in the strategic discourse, as their skills, national culture and gender affect how their work and what they can achieve is perceived (Vaara & Whittington, 2012). Perceiving ‘strategizing’ as “actions, interactions and negotiations of multiple actors and the situated practices that they draw upon in accomplishing that activity” (Jarzabkowski et al., 2007, p. 8), also implicates a psychological–communicative approach. Communication involves psychological processes to structure perceived environments, both for individual sense making and for the creation and maintenance of socially shared meaning (Hendry, 2000). Manipulation by framing can lead to information asymmetry between strategy practitioners, and consequently affect interpretations and responses to strategic decisions (Chanal & Tannery, 2011; Whittington et al., 2016).

The contribution of organizational discourse analysis in explicating the role of strategy discourse is that it highlights the constructed and enacted nature of strategy. Research on strategic discourse focuses on the social ‘decision-making process’, which actors played which roles and how their identities are constructed and the effects of the discourse of strategic outcomes.

2.1.2. Actors in strategic discourse

Strategy theories are mostly focussed on multivariate analyses of firm or industry-level effects upon firm performance, but often oversee the human actors and their interactive actions (Jarzabkowski & Spee, 2009). The discursive explanation of strategy-as-practice emphasizes the essential role of the different actors in the strategic discourse (Chanel & Tannery, 2011), and who they are goes beyond truncated views of strategy as a deliberate, top-down process. The actors are defined as strategy practitioners who shape the strategic discourse through who they are, how they act and what resources they draw upon. The actors include both those directly involved in making strategy and those with indirect influence who shape legitimate praxis and practices (Jarzabkowski & Whittington, 2008, p.101-102). This does not mean that

(17)

16 research on the top managers should be abandoned, since some valuable empirical work in SAP research indicates that there is still much to be learnt from studying these other actors as participants in strategy making rather than as its formulators (e.g. Jarzabkowski, 2003; 2005). Increasingly strategy-as-practice studies indicate that employees that lack a formal strategy role are also important strategic actors, as their actions and influence on strategy may be unintended at the firm level, they are significant for firm survival and competitive advantage (Rouleau, 2005). Additionally it is important to identify what practices provide actors outside the firm with agency in shaping strategy (Mantere, 2005). While new literature increasingly draws attention to external actors, such as non-executive directors, consultants and business gurus (Whittington et al, 2003), there remains little empirical work on how their relationships to, and engagement with the firm shape its strategy. Through a broad definition of who is a strategist, a wider range of practices, such as discourses that practitioners embody and engage in shaping strategy (Jarzabkowski and Sillince, 2007; Vaara et al, 2004; 2005) can be explored.

2.1.3. Identity construction

Organizational identity is variously defined as the way organizational members believe others see their organization (i.e., construed external image; Dutton et al., 1994), as the way that top management would like outsiders to see the organization (i.e., desired image; Whetten, Lewis, & Mischel, 1992), and as the overall impression that companies make on external actors (i.e., reputation; Bromley, 1993). As image reflects external appraisals of the organization and identity represents the perceptions of organizational insiders or strategic practitioners, Scot and Lane (2000) suggest that organizational identity and individual identities are mutually and reciprocally linked.

Scott and Lane (2000) consider this individual identity as a self-theory that “is formed and maintained through actual or imagined interpersonal agreement about what the self is

(18)

17 like” (Schlenker, 1986, p. 23 in Scott & Lane, 2000), and that it is constructed through cyclical, episodic exchanges with others, in which potential identities are “tried out” through impression management activities. These personal identities are negotiated through ongoing interaction and they necessarily draw on available social discourses on who the practitioner even can be and how he should act (Alvesson et al. 2008), creating specific expectations about the capabilities of the executive and consequently their strategic choices (Chatterjee & Hambrick, 2007; 2011; Hill et al., 2013).

Merkl-Davies and Brennan (2007) suggest that individuals themselves engage in impression management through identity creation, if it is supporting in achieving one or several strategic goals. Impression management involves constructing either public images that are a reflection of one’s positive self-concept, putting the best part of oneself into public view, or images that are inconsistent with one’s self-image through pretence (Leary and Kowalski, 1990). This self-presentational concealment is most likely to occur for individuals employed in highly visible occupations, such as executives (Whittington et al., 2016). Individuals then often strive to establish a public image that is consistent with their social role, in which they try to match their social images to prototypical characteristics fitting their role (Merkl-Davies and Brennan, 2007). In addition, executives construct images of themselves that match the values and preferences of important actors in the discourse, such by imitating the values of important stakeholder groups. Finally, identity construction also depends on executive’s current and potential image in the future, which might be based on information others are only receiving after the a substantial change has passed (Leary and Kowalski, 1990).

As practitioners are social beings, whose socio-political and rhetorical skills, and national culture and gender, not just affect how they work and what they can achieve, but also how they are perceived (Vaara & Whittington, 2012), depends the effectiveness of their

(19)

18 impression management on the publicity of the their behaviour and on the individual’s dependency on others for valued outcomes (e.g. the discretion of the executive, Hambrick & Finkelstein, 1997). This is especially relevant in the first period as executive, as they announce the most substantial changes in this period (Whittington et al., 2016).

The general note is thus that the constructed identities of an individual can be contradictory and often changing on a micro level, as they are constantly interactively claimed, contested and re-constructed in interaction with and in relation to the relevant strategic activity, the goals, the identities of the other participants and the macro social and discursive contexts (Hall, 2013; Sveningsson & Alvesson, 2003).

2.1.4. Conclusion

The theoretical approach towards strategy is based on the argument that the social world, including identities, is built from historically and contextually situated discourses (Alvesson & Kärreman 2000, Hardy 2001). Discourses are understood as specific ways of creating social reality by producing concepts, objects and subject positions that shape people’s understanding of the world and their reactions to it (Phillips & Hardy 1997). By untangling the discourses on the different levels, the interactions between the actors affecting the decision-making process can be understood and highlighted (Hendry, 2000). Additionally it allows reconstructing how framing processes and social practice have shaped strategy, actors and the decision-making process (Balogun et al., 2014). The theoretical and methodological implications of discourse analyses allows exploring to which extent a single actor has influenced the strategic choices, by reconstructing its contributions and the contributions assigned to the executive in light of other approaches (Fairclough, 2003).

(20)

19

2.2. Upper Echelons Theory

In 1984 Hambrick and Mason synthesized the fragmented literature on the characteristics of top managers into a more general ‘Upper Echelons Theory’ (UET, 1984). The central proposition of this theory is that strategic organizational outcomes, such as strategy and performance, reflect the psychological characteristics of the highest executives within an organization through the process of decision-making. By means of a theoretical framework, Hambrick and Mason (1984) portrayed the linear relationships between the main constructs; the contextual factors of external and organizational environment, the executives’ psychological characteristics, the strategic choices and the performance outcomes. The relationship between the first two constructs was substantiated with the argument that the emerging executives characteristics are in part a reflection of the contextual factors, as the characteristics are activated and shaped through information processing (Carpenter et al., 2004). In the second relationship are the executives characteristics determinants of the strategic choices, as executives make decisions and take actions that are in accord with their personal preferences and biases. Through these choices, the characteristics affect and are reflected in the organizational performance (Hambrick & Mason, 1984).

To explain how the strategic choices reflect the executive psychological characteristics, the UET refers to the concept of bounded rationality affecting strategic decision-making (Hambrick & Mason, 1984; Simon, 1991). The work on bounded rationality suggests, “informationally complex, uncertain situations are not objectively "knowable" but, rather, are merely interpretable” (Mischel, 1977, in Hambrick, 2007, p. 334). The UET theorizes how executives’ psychological characteristics influence their field of vision, selective perception and interpretations, to make sense of the overwhelming amount of environmental stimuli they face (Rost & Osterloh, 2010). In turn this personalized reality affects their strategic choices and the performance outcomes (Hambrick, 2007, p. 334). This

(21)

20 rational approach towards the decision making process explains how different executives can examine the same strategic situation and come to different conclusions about how to respond (Cannella & Pettigrew, 2001; Cannella & Holcomb, 2005).

While much empirical research has been conducted in support of the UET, three facets signiﬁcantly challenged the body of work (Rost & Osterloh, 2010): The multi-directional nature of the Upper Echelons Theory, the focus on main influencer and the measuring of executives’ characteristics.

2.2.1. Complex UET model

In their original model, Hambrick and Mason (1984) emphasized a rational, linear, progressive relationship between the main constructs (Eggers & Kaplan, 2013). At the same time, they conceded that causality surrounding the UET model was muddied and reverse logics might be at play, as the three main constructs all interact in determining organization performance levels (1989, p. 197; Hambrick, 2007). More recent research confirmed that the top executives and their activities fail to occur in a vacuum (Carpenter et al., 2004).

Firm outcomes and context, including environmental and industry contingencies, play a critical role in determining who serves among the firm’s top executives and thus which characteristics are needed or will emerge (Carpenter et al., 2004). Furthermore executives are also selected to fulfil a certain assignment or follow a certain path, thus their strategic choices are also due to their mandate, additionally to any unwittingly biased information processing (Certo et al., 2006). The influence of the single executive on the organization is also affected by intervening variables such as managerial discretion (Crossland & Hambrick, 2011; Finkelstein & Hambrick, 1990) and high job demands that put pressure on the individual (Hambrick et al., 2005; Hambrick, 2007). Blettner et al. (2012) concluded in their quantitative research on the CEO complex fit perspective, that the individual effect of an executive in a

(22)

21 certain firm is contingent to some degree on various unique combinations and interactions of environmental events, trends, and forces, and on the internal configuration of the firm.

In response to these remarks are the complex and iterative relations between the main constructs of the UET reduced to a simplified multi-directional model below (see Model 1). The relations between the constructs are perceived as correlating, without clear causal direction as research delivered contradictory results (Blettner et al., 2012; Carpenter et al., 2004; Wang et al., 2016).

2.2.2. Actors of influence in the UET

After the first article on the Upper Echelons Theory (Hambrick & Mason, 1984), the researchers acknowledged that management is a shared effort in which a dominant coalition collectively shapes organizational outcomes. They suggested that a focus on the characteristics, the collective cognitions, capabilities, and interactions of the entire the top

Model 2. Simplified Upper Echelons Theoretical model

(23)

22 management team (TMT) would yield better explanations of organizational outcomes than the customary focus on the individual top executive alone (Hambrick, 2007).

While a considerable amount of researchers use the model to describe team-level decision making (for an overview see Certo, Lester, Dalton, & Dalton, 2006), multiple studies revealed that the model is still inherently individual-level in focus, and several important limitations must be overcome before the model will provide a full explanation of team-level decision making (Carpenter et al., 2004; Nadkarni & Herrmann, 2010; Wang et al., 2016). They described two major shortcomings in the TMT approach (Zhou, 2016). First, TMT’s are often perceived as heterogeneous and their team demographics (e.g. age, gender) are assumed to reflect team deep level composition such as personalities and attitudes (Pitcher & Smith, 2001) and second, the mediating team process variables, such as leadership and conflict, are often overlooked (Cannella & Holcomb, 2005). This has led to far stretched correlations between individual characteristics, strategic choices and organizational outcomes (Cannella & Holcomb, 2005).

Yet most researchers do acknowledge the relevance of a dominant coalition that collectively shapes organizational outcomes and affects the influence of the individual executive. The board of commissioners and other managers comprising the internal labour market can exert an indirect influence on organizational strategy and performance, through executive selection and communication with the media (Wang et al., 2016). The theoretical understanding of external constraints has also expanded to the influence of stakeholder interests, institutional forces through media and politics, and broader factors such as environmental uncertainty (Carpenter et al., 2004). Researchers’ focus should thus not be exclusively internal, but also see how other actors indirectly affect the executive and organizational outcomes (Cannella & Holcomb, 2005).

(24)

23 2.2.3. Executives’ characteristics

Hambrick and Mason (1984) acknowledged in their article the methodological challenges that came with the central role of critical concepts such as psychological characteristics. These illusive concepts were perceived difficult to be measured directly, as executives are often reluctant to participate in research (Chatterjee & Hambrick, 2007; Hill, White & Wallace, 2013). Hambrick and Mason (1984) proposed the use of observable characteristics of executives as valid, but incomplete, indicators of executives’ psychological characteristics, as they are easily available, reliable and accessible (Cannella & Pettigrew, 2001; Lawrence, 1997). Observable demographic variables can prove good predictors and reasonable substitutes for subjective concepts as long as they show high reliability and validity (Lawrence, 1997).

Though the theoretical argumentation of this approach developed valid predictions of strategic decisions (Carpenter et al., 2004; Hambrick, 2007), empirical quantitative research on these variables produced mixed findings. For example, researchers found that the tenure of a chief executive is positively (McClelland, Liang, & Barker, 2010), negatively (Nadkarni & Hermann, 2010), or not related (Balkin, Markman & Gomez-Mejia, 2000) to firm performance. Additionally a lack of empirical support for the real underlying causal processes related to the role executives’ psychological characteristics in decision-making remained (Cannella & Pettigrew, 2001; Carpenter et al., 2004; Hambrick, 2007; König et al., 2014).

Research on psychological characteristics outside the UET continued and conceptualized different frameworks to collect and approach these illusive concepts. Through methods of self-assessment they comprised the contents of almost every major personality inventory into the six-factor HEXACO personality framework (Wang et al., 2016). But after comparing the methods to assess these characteristics, Boyd et al. (2015) found that self-report questionnaires on these abstract and complex phenomena are inadequate for painting an

(25)

24 accurate picture of individual mental life. They suggested analysing already available data, collected through unobtrusive methods (Boyd et al., 2015). This is an approach that allows researchers to uncover the underlying processes without the cooperation of a respondent or contaminated response (Chatterjee & Hambrick, 2011; Hill et al., 2013).

Simultaneously the use of unobtrusive indicators for personality factors made its introduction in UET research as most promising alternative to empirically assess the psychological characteristics (Chatterjee & Hambrick, 2011). For example, Hambrick and Chatterjee (2007) concluded from using unobtrusive measures that the narcissism personality factor of CEOs is positively related to strategic dynamism and grandiosity, as well as the number and size of acquisitions, and it engenders extreme and fluctuating organizational performance. O’Reilly et al. (2014) showed that CEO personality affects a firm’s culture and that culture is subsequently related to a broad set of organizational outcomes including a firm’s financial performance, reputation, analysts’ stock recommendations, and employee attitudes. And the work of Peterson et al. (2003) provided dramatic evidence of the impact of executives’ personality on firm performance through their effect on TMT dynamics. More recently the fine-grained UET analyses of Wang et al. (2016) based on the HEXACO framework show that the personality factor extraversion among CEOs was positively related to firm strategic actions and future firm profitability. However, given the relatively small number of studies involved in the meta-analyses, more future research on the relationships of CEO personalities with firm strategic actions and performance was suggested (Wang et al., 2016).

2.2.4. Changing perspective on the UET

Several meta-analyses reflect the recursive effects, problems of endogeneity and the multi-directional nature of the Upper Echelons Theory and describe the need for more comprehensive and dynamic approaches (Blettner, Chaddad & Bettis, 2012; Carpenter et al.,

(26)

25 2004; Wang et al., 2016). In response Hambrick (2007) suggested considering executive psychological characteristics as consequences rather than as causes for strategic choices and organizational outcomes and an additional focus on the non-instrumental phenomena that surround executives by means of qualitative research and a combination of disciplines. This might improve predictions of how and why executives' characteristics become manifested in strategy (Hambrick, 2007). In line with this suggestion König et al. (2014) proposed to take on a different perspective to 'open the black box' of unknown, underlying psychological and social processes of individuals and organizations, by analysing discourses.

(27)

26

2.3. Converging theories

The mutually constituting construction of the social reality in discourses translated in the oscillating pattern between different levels of analysis shows resemblance to the complex relationship between constructs in the UET (Balogun et al., 2014). In the UET model other modes, different from the internal rational decision-making process, by which executives influence organizational outcomes are overlooked (Carpenter et al., 2004). Decision-making is but one of several roles assumed by senior leaders. Recent developments suggested that executive attention to institutional constraints might directly affect organizational strategy and performance. Indirect effects are observed in relation to socially constructed, normative standards (Carpenter et al., 2004). Thus, rather than reflect the decision processes outlined in the UET (Hambrick & Mason, 1984), executive effects may reflect the firm’s and its management’s embeddedness in a broader institutional environment. This suggestion is in line with the discursive view on strategy and further explored below.

The integration with the discursive approach perceives the theory in a different light as presented below (Model 3, Integrated UET discourse model). The environmental factors show resemblance to what is described as the macro contextual level factors, not just as antecedents to selective perception but also as explanation for strategic actions and changes. The process of strategic practice and interaction, in which texts are produced and interpreted is where the strategic decision-making takes place, interpreted as social constructed activity. The different actors facilitate and take part in the process by using the micro level constructs to frame the interpretations of events. On the micro textual level are the two constructs represented that tie the constructed personalities of the individual executive and the constructed strategic choices together.

The remaining review contains the exploration of the relevant findings on the main constructs, the contextual factors of external and organizational environment, the process of

(28)

27 strategic decision making, the strategic choices and the executives psychological characteristics, in light of both directions.

2.3.1. Contextual environmental factors

In the original theory the primary construct portrayed is the ‘objective situation’ the organization faces (Hambrick, 1984). The theory suggests that the executives’ characteristics are in part a reflection of this situation, which encompasses all potential environmental and organizational stimuli (Hambrick, 1984). The role of the contextual factors as antecedent, mediator and moderator for the predictive nature of upper echelons psychological

Model 3. Integrated model

(29)

28 characteristics is largely represented in theoretical models, though remained scarcely explored (Carpenter et al., 2004; Quigley & Hambrick, 2015; Yamak et al., 2013).

The conceptualization of the contextual constructs in the UET varies between the

external environment, e.g. stability and change in environment, industry and institutional

characteristics (Shepherd et al., 2014; Yamak et al, 2013) and the organizational environment, e.g. changes in firm and board characteristics (Carpenter et al., 2004). These contextual factors correspond with the macro level themes and trends in organizational discourse (Phillips & Oswick, 2012). At the macro-level, discursive work tends to draw heavily on situational, societal and institutional narratives that interact with the underlying strategic discursive practices, such as ‘feminism’ and ‘globalization’ (Ruiz, 2009).

The external environment is conceptualized as the settings in which firms compete and

make strategic choices (Yamak et al., 2013) and is characterized by the considerable attention to stability and change (Carpenter et al., 2004; Finkelstein et al., 2009) There appears to be general consensus that environmental characteristics, particularly those that represent uncertainty for the firm and its managers, will have implications for the UET (Carpenter et al., 2004). High uncertainty industries are able to afford executives with greater discretion (Finkelstein & Hambrick, 1990), which increases their individual effect on performance (Hambrick & Quigley, 2013). However research has revealed that under conditions of uncertainty information processing is worse among male managers and financial experts, negatively affecting their ability to predict future scenarios (Rost & Osterloh, 2010).

The contextual construct of the external environment ‘Institutions’ directly reflects the institutional themes in organizational discourse analysis (Phillips & Oswick, 2012). While much of the UET research examines the effects of institutions on organizations, discourse analysis can be used theoretically to explicate institutional processes, such as theorization and translation, which are fundamentally affecting the social construction of concepts. Discourse

(30)

29 analysis also can be used to understand how the institutional logic that characterized a field at a moment in time came to be (Phillips & Oswick, 2012).

The organizational environment in the UET is conceptualized as the internal firm and

board settings in which executives operate, and is characterized by changes in strategy and structure (Carpenter et al., 2004; Shephard & Rudd, 2014). For example individuals outside of the top executive team, such as the board of directors, may impact the UET model as they exert a direct influence on organizational strategy and performance, through the provision of important information, legitimacy, and other resources (e.g., Geletkanycz & Hambrick, 1997). At the same time, they can exert an indirect effect by counsel and advice and through selection of executives (Carpenter et al., 2004).

Connecting organizational discourse analysis to changes in organizational internal context highlights how the production and dissemination of texts influence the way in which organizational change takes place (Phillips & Oswick, 2012). Heracleous and Barrett (2001) performed a discourse approach combining both action and structure. As they looked at central themes and intertextual discourses facilitating or hindering organizational change, they found discursive shifts at both the communicative action and deep structure levels. Their research reflected discursive conflict between stakeholder groups concerning the change process. Organizational discourse analysis provides the tools to explore the processes of the construction and reconstruction of meaning that underlie change (Phillips & Oswick, 2012) and takes this primary UET construct into account as explanation for how discursive practices and strategic decision making takes shape.

2.3.2. Strategic decision-making

To explain how the strategic choices reflect the executive psychological characteristics, the UET is refers to the concept of bounded rationality (Hambrick & Mason, 1984; Simon, 1991). This traditional perspective on decision-making suggests that executives subconsciously

(31)

30 develop cognitive filtering mechanisms to deal with the continuous complex and ambiguous overload of information they are faced with. These filtering mechanisms are structured mental models to decide on the selection and relevance of information and whether it provides primary context for decisive acts of implementation (Hendry, 2000). What and how temporal mental models emerge, relies on the engrained psychological characteristics such as personality factors (Kahneman, 2003). The characteristics affect intuition and the accessibility of thoughts (Kahneman, 2003). Additionally the contextual factors providing information affect the priming of mental models (Kahneman, 2003). The theory thus suggests that executives’ psychological characteristics affect what information is processed into strategic decisions (Hambrick & Mason, 1984).

A critique on the perspective is that the process is not empirically proven and thus pre-empts discussion as to how decisions and actions are related (Eggers & Kaplan, 2013; Hendry, 2000). As the initially linear theory on decision-making evolved, the factor of information framing was also acknowledged. Framing is “a [constructed] lens, through which actors reduce the complexity of the environment in order to be able to focus on particular features, make interpretations, [or] decide to act" (Kaplan & Tripsas, 2008, p.791). The basic principles of framing are that the mental models are communicated through the representation of situations and that the formulation is passively accepted by the receiver. The acknowledgement of this factor has several implications. First, it enables researchers to empirically perceive the projected mental model of the sender, who expressed the information and thus understand how the information can be received. Furthermore, it acknowledges that the availability of mental models can be manipulated externally and thus that the decision-making process isn’t limited to a mental process and involves multiple actors.

Hendry (2000) developed an empirically grounded conceptualization of strategic decision-making as elements of a strategic discourse, in which decisions may also be created

(32)

31 in advance of any commitment to action. He conceptualized strategic discourse as “not only the medium in which decisions are discussed and recorded, but also the medium through which interpretations are developed and expressed and strategic actions initiated, authorized and acknowledged” (Hendry, 2000, p. 957). This integrating conceptualization of the traditional, action and interpretative perspectives on strategic decision-making, allowed Hendry (2000) to distinct three core elements: strategic intentions (e.g., ‘Intentional decisions, as identified retrospectively by actors and researchers’, Hendry, 2000; ‘idea or talk’, Vaara, et al., 2004), interpretations or expectations (e.g., decision statements as outcomes of cognitive processes’, Hendry, 2000; ‘stakeholder perceptions of future strategic choices and conditions’, Whittington et al., 2016; ‘how stakeholders ‘construct’ strategic choices’, Chanal & Tannery, 2011), and actions (e.g., ‘Decisive acts, commitment of resources’, Hendry, 2000; ‘practice and action’, Vaara et al., 2004; Wang et al., 2016).

Strategic intentions can originate from changes within industry or the organization, earlier chosen strategies or the executive. The macro contextual environmental factors, such as to political instability or changes in TMT composition can lead to divestments or reorganization (Blettner et al., 2012). Earlier chosen strategies, such as acquisitions or mergers, can lead to intentions towards organizational change or repetition of choices. The intention might also cause resistance to change by employees, which can affect the continuation of the strategic choice (Helfat & Martin, 2015). The public expression of strategic intentional decisions by the chief executive or organization also initiates the socially constructed decision-making process and allows interpretation by stakeholders (Chanal & Tannery, 2011; Whittington et al., 2016). Competitors can anticipate on announced strategic actions, by initiating a counter strategy (e.g., competitive signals, Porter, 2008). The industry, another component of the external environment, might respond in their valuation of the

(33)

32 organization as financial analysts evaluate intended strategic decisions based on the perceived identity of the executive (Graham, Harvey, & Rajgopal, 2005).

The perception of the strategic decision-making process as a social constructed process instead of the cognitive bounded rationality process, allows researchers to explain the causality surrounding the UET model, as the discursive explanation takes the iterative relations between the main constructs into account.

2.3.3. Strategic choices

The traditional rational perspective on decision-making takes strategic decisions as the ‘decisive’ elements of an empirically observed rational process. At a point in the history of a strategic change a decision is made that commit the organization to proceed in a certain way. Such decision points may not always be clearly distinguishable in the historical record, as the processes involved may be complex, iterative and multi-layered. However this does not, on the traditional rational view, detract from their importance and centrality to the strategy process. Within the body of Upper Echelons research a variation of concepts are used to describe strategic organizational outcomes. In the original theory a difference was made between ‘strategic choices’ and the result from these choices, ‘performance’ (Hambrick & Mason, 1984). The most recent meta-analysis within the UET conceptualized the construct of strategic choices as expensive investments that alter the organizations’ scope, involve significant risk, and are used to make changes or adjustments to firm strategy (Wang et al., 2016).

Strategic scope is conceptualized as the composition of geographic and product

markets in which an organization competes (Holcomb, Holmes, & Hitt, 2006). Typical strategies are acquisitions, divestments, international diversifications and product diversifications (Wang et al., 2016). Research provides evidence that companies initiate more mergers and acquisitions when their chief executive is more risk-tolerant (Graham et al.,

(34)

33 2013). International operations are risky because they expose firms to multiple environments, many of which are difficult to predict and understand, e.g., emerging economies (Wang et al., 2016). As international acquisitions often require substantial resources to establish and manage, they significantly influence firm performance (Herrmann & Datta, 2002). Divesting often reduce risk and uncertainty, as they eliminate business units, but can be an indication of low performance (Eisenmann, 2002).

The second strategic construct, strategic risk, refers to “research and development (R&D) expenditures, capital investment, and long-term debt” (Devers, McNamara, Wiseman, & Arrfelt, 2008, p. 550). Given that advertising intensity, percentage of total sales, captures firm’s strategic differentiation and large-scale irreversible resource commitments bear with uncertainty, these are characteristics for strategic risk. Hambrick and Quigley (2013) concluded that in high advertising intensity industries the chief executive can deliver a bigger impact over industries with low advertising intensity.

The last strategic construct is strategic change, which refers to “a difference in the form, quality, or state over time in an organization’s alignment with its external environment” (Rajagopalan & Spreitzer, 1997, p. 49). By changing strategy, new routines and processes are implemented to manage the change. As this alters organizational structure, it is difficult to reverse and often results into employee resistance to change (Wang et al., 2016). Research has shown that executives who are more effective at task-oriented behaviours are more likely to focus on both the mobilizing and evaluating activities associated with organizational change implementation, and executives who are more effective at person-oriented behaviours are more likely to focus on the communicating activities of organizational change implementation (Battilana et al., 2010).

From a discursive perspective strategy-as-practice theorists suggest instead of viewing strategy as a unified body of knowledge or hegemony, to maintain that it is important to focus

(35)

34 attention on the various alternative discursive practices that enable and constrain organizational actors as practitioners of strategy (Vaara, 2010). These strategic choices in the UET are thus in line with this approach also perceived as a situated socially accomplished activity. The approach acknowledges the differences made between the strategic choices, as Seidl (2007) suggested that strategy should be conceptualized not as a unified body but rather as fragmented into a multitude of autonomous discourses that take very specific forms in organizations. Boje (2008) has in turn argued that strategy making deals with a variety of narratives that may co-exist in organizations. As different actors take place in discourse it is also suggested that they all can initiate the strategic discourses.

Strategic choices, which can entail scope, risk or change in the organization, are not just actions made by the organization, but also suggestions in the shape of intentions, expectations and interpretations by the media, stakeholders and the executives.

2.3.4. Psychological characteristics

The organizational literature is dominated by concepts of strategy as a top-down process of formulation separated from implementation, predisposing a focus upon top managers, their demographics and their decision-making processes (e.g. Hambrick and Mason, 1984). But this dominant definition of strategists is inadequate, as the demographics such as age, tenure, educational and functional background, ethnicity and gender were just proxies for behaviour, instead of a starting point from which to study actual behaviour (see also Cannella & Pettigrew, 2001; Carpenter et al., 2004). These approaches have ignored the individual experiences of agency, in which who a person is, is inherently connected to how that person acts and the consequences of that action. For example, how strategy is defined and is affected by the identity of the individual, which can be explored by discourse analysis.

UET research on individual personality constructs that relate to a person’s positive self-concept has recently made its introduction, challenging the use of proxies (Carpenter et

(36)

35 al., 2004; Wang et al., 2016). Personality is a “relatively permanent, ingrained disposition” that affects how executives attend to and process information about the environment, the firm, and their own capabilities (Finkelstein et al., 2009, p. 70). The personality construct covers similar mental processes as the original cognitive base and values, but allows easier procurement of empirical data through unobtrusive methods (Peterson et al., 2003). Chatterjee and Hambrick (2007) for example measured the effect of narcissism of chief executive officers on a firm’s strategy and performance through the prominence of the CEO’s photograph in annual reports, the prominence in press releases and use of first-person singular pronouns in interviews.

Wang et al. (2016) combined the different articles within the UET that covered partial constructs of what is called the HEXACO Personality Inventory (Ashton & Lee, 2005, 2007) and compared it with the strategic choices made by executives. The HEXACO Personality Inventory (HEXACO-PI) consists of 24 facet-level personality trait scales that define the six personality factors named Honesty-Humility, Emotionality, Extraversion, Agreeableness,

Conscientiousness, and Openness to Experience. Wang et al. (2016) concluded that executive humility, emotionality and extraversion were all significantly related to organizational strategic actions, but executive conscientiousness was not. Specifically, executive’s grandiose self-concept (i.e., lack of humility) was positively related to strategic risk as well as

organizational risk taking and acquisition (Wang et al., 2016). Agreeableness was due to an insufficient number of primary studies not analysed (Wang et al., 2016). This framework of personalities was aimed to categorize the variety of cognitive processes, values and behaviour into six dimensions through self-evaluation methods, allowing researchers to perform factor analyses with these illusive processes (Lee & Ashton, 2004).

This perspective on the psychological characteristics of the executive also allows a qualitative approach, in which “personality is an abstract category we use to express a holistic

(37)

36 sense of how we interpret a person’s identity and emotional constitution” (Joseph, 2004, p. 85). In this sense personality is not just how the individual perceives him or herself, rather it is viewed as socially constituted and a reflexive, dynamic product of the social, historical and political contexts of an individual’s lived experiences (Hall, 2013, p. 31). This discursive approach to personality highlights its temporary, context-sensitive and evolving nature and is in contrast to the essentialist view of identity as the set of fixed characteristics of a person (Alvesson et al. 2008). So what is involved in discourse analysis is not a rejection of all attributions of psychological or social attributes to people, but rather an insistence that only certain kinds of attributes are relevant. In effect, discourse analysis treats attribution of the personalities deriving from recent processes of discursive interaction as legitimate, while rejecting any produced by genetic constitution or by early upbringing (Hammersley, 2003).

Based on the idea that individuals constantly strive to shape and change their personal identities in organizations and are being shaped by discursive forces, researchers have explored the constructions of various managerial and non-managerial identities in a complex, changing and multi-ordered organization (Sveningsson & Alvesson, 2003; Vaara & Whittington, 2012). Managers influence stakeholder identification by presenting desired organizational images that are attractive to stakeholders and by influencing the extent to which stakeholders' participation in the organization attracts and holds their mental attention (Scott & Lane, 2000). But also in the process of decision-making, at both the presentation of the intentions and the interpretation of strategic choices are psychological characteristics at play through framing (Cannella & Holcomb, 2005).

A key assumption is that an executive or organization primarily gives sense to information, by communicating strategic intention, in the context of analyst evaluations (König, et al., 2013). These evaluations are partly based on the perceived constructed identity of the main strategic decision maker. Analysts make sense of this information with the

(38)

37 explicit goal of giving sense to other stakeholders, particularly investors, which in turn affect their appreciation of the organization and industry as they discursively negotiate meanings (Anderson, 2005). In this process of sense giving actors "frame" the firm's current and future situation by means of the construction of an identity.

2.4. Conflicting paradigms

Although discourse analysis has become increasingly common in many areas of organization and management research (Phillips et al., 2008), the discursive approach and the UET have hardly met in research. This lack of previous interaction between two approaches with these large similarities can be found in their original paradigms, as Vaara et al, (2004, p.1) conclude: “Strategy research has paid little attention to the discursive processes involved in strategizing. Many scholars have probably felt that the social construction of reality is outside the core of strategy research and should be left to the sociologists. Others have viewed the rhetorical or discursive as interesting side issues, but not as important as the ‘‘real’’ processes involved in strategizing”. This highlights the difference in assumptions and beliefs researchers bring to their research work with regard to questions of ontology and epistemology (Guba & Lincoln, 1994). As the methods of the UET and discourse analysis are determined by these questions, they both present relevant, but conflicting paradigms.

The original Upper Echelons Theory appears to result from a form psychological behaviourism (Carpenter et al., 2003). Behaviourism is a paradigm that emphasizes the ‘black box’ model, which implies that researchers have to rely on perceptible behaviour of people, because internal processes can’t be measured (Burck, 2005). A weakness of psychological behaviourism is a tendency to treat anything that is not observable in the defined sense as having a different ontological status from that which is (Hammersley, 2003). The effect of this is that researchers have to operate on the basis of a very thin model of the human actor one whose concerns are exhausted by what is ‘observable’ (Hammersley, 2003).

(39)

38 This is in direct contrast with the constructivist ontology that is at the core of the discursive approach. Social constructivism considers research data, such as the accounts of research participants, as ‘constructed’ within a particular research context, rather than as an objective reflection of ‘reality’ (Burck, 2005). It draws on the idea that our ways of knowing are negotiated through social interactions over time and in relation to social structures, contexts and resources which support or indeed suppress these ways of knowing (Shotter, 1993). It tends to reduce everything to flows and the constant state of movement and flux, which they generate and reproduce (Harvey, 1998). There isn’t one truth, but there exist plenty of truths beside each other. The subjectivity of the human perception and judgement is the central ontology (Simons, 2000).

The reason these two paradigms did not meet is that in discourse analysis the attribution of substantive psychosocial characteristics to people had to be avoided because those characteristics are discursive products rather than ontologically given properties of the people concerned (Hammersley, 2003). To overcome this the work in discourse analysis have developed an ‘discursive social psychology’ approach. This as ‘an approach to social psychology that takes the action orientated and reality-constructing features of discourse as fundamental’, rather than being ‘a social psychology of language’ (Potter & Edwards, 2001, p.2). Thus, personalities are treated not as ‘inner entities that drive behaviour’, but as evaluations that are part of discourse practices. Personalities are seen as constituted in and through participants’ ways of talking. A change in the view about the ontological status of social phenomena can thus overcome the discrepancy between the UET and discourse analysis. By treating them as discursive products the focus of inquiry becomes how and why they are constructed in the way they are (Hammersley, 2003).

(40)

39

2.5. Conclusion

By means of this literature review and the introduction of discourse to the Upper Echelons Theory, it is theoretically proposed that two of the main constructs in the UET, personality and strategic decision-making are equally ‘illusive cognitive processes’ as much as that they exist ‘by social construction in communication’. This has methodological implications, as they can be empirically perceived by reconstruction of the emerging discourses though unobtrusive measures, which are already applied to uncover single personality factors. This methodological approach implies a way to move beyond the marginally used self-evaluations (Hambrick & Mason, 1984), and perceive these illusive constructs through the eyes of internal and external actors, This approach acknowledges the participation of other actors in the decision making process, such as top management team members, board members, media, analysts, competitors and the influence of other contextual factors on how strategic choices are constructed.

Perceiving personality factors, identities and strategic choices as socially constructed discourse analysis created the opportunity to analyse how the influence of executives, particularly chief executive officers, is attributed. How actors negotiate on the role and identity of the upper echelons is relevant, as the portrayed and perceived personality, position and track record of the executive affects interpretation, expectation and anticipation of the strategic choices (Mantere & Vaara, 2008; Vaara & Whittington, 2012).

In the eyes of the beholder : let's talk strategy and identity interpretation : a research on the discursive construction of executive personality factors and their influence on strategic choices