More specifically, the upper echelons perspective encapsulates three subordinate ideas (Carpenter, Geletkanycz & Sanders, 2004; Hambrick, 2007), discussed in detail below: (1) senior managers’ cognitive base and values become reflected in strategic outcomes; (2) observable demographic characteristics are reliable indicators of executives’ cognitive frames and, as a result, can be used to predict strategic outcomes; (3) studying the characteristics of a firm’s upper echelons as a whole (i.e.,  entire top management team) yields stronger predictions of strategic outcomes than focusing on the chief executive officer (CEO) alone.

The upper echelons logic of strategic choice

The first tenet of UET builds upon the premise that strategic situations encapsulate far more stimuli than decision-makers can comprehend (Cyert & March, 1992). Therefore, UET posits that managers try to interpret strategic issues and devise alternative courses of action by taking mental shortcuts and relying on their previous experiences (Hambrick, 2018). For instance, when faced with an unprecedented environmental shock, managers may utilise their previous experience when handling other types of business crises in order to analyse the situation and develop an appropriate response and a set of actions. The underlying perceptual process is delineated in Figure 2.

Each business manager carries their own cognitive base and set of values (e.g., experiences and personality) to the decision-making process, which serve as a way of filtering strategic situations. As illustrated in Figure 1, the cognitive base and values “create a screen between the situation and the eventual perception of it” (Hambrick & Mason, 1984: p195). There are three sequential mechanisms through which this “perceptual screen” operates: (i) limited field of vision, whereby managers’ personal biases pose a sharp limitation on the environmental and/or organisational stimuli to which attention is directed; (ii) selective perception, which implies that managers have limited information processing capacity, and therefore analyse only some of the phenomena encompassed in their field of vision; and (iii) interpretation, whereby managers utilise their cognitive frames to interpret the selected phenomena. Eventually, managerial perceptions of the actual strategic situation provide the basis for strategic choice.

Observable managerial characteristics as proxy indicators of cognitive base and values

The second main tenet of the UET is anchored in organisational demography (Pfeffer, 1985). It suggests that managers’ cognitive base and values are reflected in observable characteristics such as age, education, functional background, and other career experiences such as organisational tenure and aspirations (Wiersema & Bantel, 1992). Admittedly, there are several difficulties in obtaining psychometric data on top executives’ cognition, values, and perceptions (Nielsen, 2009). Unobservable, psychological constructs are not convenient to measure and, at times, are not even amenable to direct measurement (Hambrick & Mason, 1984). Considering also that an individual’s cognition is shaped by their background characteristics and life experiences, demographic variables are thought to be valid and reliable indicators of the psychological processes that shape how managers interpret strategic situations and formulate appropriate strategic alternatives (Carpenter, Geletkanycz & Sanders, 2004). In fact, a plethora of empirical studies has demonstrated that executives’ demographic profiles are strong predictors of strategic choices and performance outcomes (Hambrick, Cho & Chen, 1996; Boeker, 1997; Ferrier, 2001; Carpenter, 2002). Although this tenet does not account for the actual processes that drive executive behaviour (Lawrence, 1997; Priem, Lyon & Dess, 1999), it ensures the reproducibility of empirical findings and facilitates the genesis of an ongoing research program. As Weick (1979) states, empirical research can become more cumulative if theoretical concepts can be defined in terms of observable indicators.

UET emphasizes seven key demographic variables that can be used to predict strategic outcomes: age, functional background, career experiences, education, socioeconomic background, financial position, and the heterogeneity of these characteristics within a firm’s top management team (TMT). Age indicates executives’ receptivity to change and willingness to take risks (Hambrick & Mason, 1984). Compared to older managers, young managers tend to pursue more risky choices, such as strategic change (Wiersema & Bantel, 1992) and significant investments in research and development (R&D) (Barker & Mueller, 2002). Functional background plays a central role in strategic decision-making as individuals working in different functional areas develop distinct perceptions about a firm’s strategic goals (Geletkanycz & Black, 2001). Managers that have accrued their experience mainly from throughput functions, such as production and accounting, place emphasis on efficiency-related issues, whereas managers with more experience in output functions, such as marketing and R&D, favour innovative strategies that can enable business growth (Barker & Mueller, 2002).

UET further argues that executives with different career experiences, such as organisational tenure (i.e., the length of time an executive has worked for a specific organisation) and  industry or organisational experience (i.e., the different types of industries or organisations an executive has worked for), differ in their strategic choices, due to their exposure to diverging perspectives and environments (Hambrick & Mason, 1984). For instance, chief executives that are new to an organisation tend to make more strategic changes as they are less committed to the status quo and bring new perspectives into the organisation (Boeker, 1997). Longer-tenured executives tend to be more attached to an organisation, and, as such, exhibit emotional and/or political resistance to change (Hambrick, Geletkanycz & Fredrickson, 1993). In a different vein, a decision maker’s educational background is regarded as an indicator of cognitive ability and skills (Wiersema & Bantel, 1992). Higher levels of education have been associated with an enhanced ability to process information (Hambrick & Mason, 1984), creatively deal with complex administrative situations (Bantel & Jackson, 1989), and tolerate ambiguity (Dollinger, 1984). Hence, highly educated managers are more likely to pursue innovative strategies (Kimberly & Evanisko, 1981) and exhibit greater awareness of and receptivity to the need for strategic change (Wiersema & Bantel, 1992).

According to the upper echelons perspective, socioeconomic background and financial position characteristics can also affect decision-makers' choices. Managers from lower socioeconomic groups seek greater recognition and esteem through their actions (Hambrick & Mason, 1984). As such, firms whose top managers come from relatively disadvantaged backgrounds tend to be more aggressive in their strategic actions and exhibit higher levels of acquisition activity and unrelated diversification (Channon, 1979). Top managers’ financial position, which refers to the extent of stock ownership and total compensation, is thought to be an indicator of their inclination towards short-term versus long-term, highly rewarding / profitable actions (Luo, Wieseke & Homburg, 2012). Managers with substantial stockholdings are more committed to the maximisation of shareholder wealth and undertake actions that pay off in the long run, such as R&D investments (Barker & Mueller, 2002). However, managers without significant wealth at risk focus on current profitability and avoid risky investments, yet highly rewarded by the stock market (MAY, 1995).

Finally, UET posits that TMT heterogeneity, or the amount of dispersion within a managerial group regarding members’ characteristics, is highly pertinent to the study of strategic decision-making, as it represents the diversity of a team’s cognitive base and values (Finkelstein, Cannella & Hambrick, 1996). Heterogeneous TMTs encapsulate divergent perspectives, expertise, and knowledge bases, enhancing decision-making quality, especially when encountering ill-defined and novel situations (Nielsen, 2009). Diverse teams are willing to challenge each other’s viewpoints and be more comprehensive when making strategic decisions (Simons, Pelled & Smith, 1999). Diverse teams also show high levels of creativity and innovativeness (Bantel & Jackson, 1989). On the other hand, TMT heterogeneity can be associated with inferior decision-making (Hambrick, Cho & Chen, 1996). At high levels of diversity, conflict is more likely to occur, which in turn, leads to a low group consensus on strategic choices (Knight et al., 1999).

The Top Management Team (TMT) as the unit of analysis

The third subordinate idea introduced by Hambrick and Mason (1984) posits that studying entire top management teams instead of individual chief executives yields stronger predictions of organisational outcomes. Chief executives typically share decision-making responsibilities and power with other members of the TMT. Therefore, the cognitive frames of the entire team enter into the process of interpreting and responding to strategic situations (Hambrick, 2007). If two firms have CEOs exhibiting similar characteristics while their management teams consist of executives with highly distinctive backgrounds, studying the entire team would improve confidence in predicting the two firms’ strategies. This is consistent with empirical research showing that TMT characteristics matter more in decision-making than CEO characteristics alone (Papadakis & Barwise, 2002).

Theory Updates/Extensions

Moderators of the upper echelons logic

Since the seminal postulation of UET (Hambrick & Mason, 1984), a substantial stream of research has focused on establishing the boundary conditions of UET. Scholars have identified various moderators of the relationship between upper-echelon characteristics and strategic outcomes, including managerial discretion (UET1), TMT structure (Hambrick, 1995; Hambrick, Humphrey & Gupta, 2015), executive job demands (Hambrick, Finkelstein & Mooney, 2005), and managerial power (Finkelstein, 1992). According to this stream of research, UET has greater predictive strength in some contexts than in others (Figure 3).

The first and most notable refinement of UET has been the introduction of “managerial discretion” as a moderator of the upper echelons logic (UET1). Although UET posits that strategic outcomes reflect executives’ characteristics, it cannot be neglected that some executives have greater control over what happens in their companies than others (Haleblian & Finkelstein, 1993). Managers with little or no control would not be able to influence their organisations’ strategies to the same degree as managers with high levels of discretion. Thus, strategic decisions and performance outcomes can be predicted by upper-echelon characteristics depending on the extent to which they have a “latitude of action” (Finkelstein & Hambrick, 1990). If managers have great discretion, their characteristics will be highly reflected in organisational outcomes. However, managerial characteristics cannot predict organisational outcomes if discretion is lacking. Empirical research has consistently supported the importance of managerial discretion as a boundary condition of UET. The effect of CEO characteristics on organisational performance was found to be significantly stronger in high-discretion (e.g., US) than in low-discretion (e.g., Japan) national contexts (Crossland & Hambrick, 2011). Also, TMT characteristics have a greater influence on strategic change decisions in high-discretion industries (e.g., computer industry) than in low-discretion ones (e.g., natural gas industry) (Finkelstein & Hambrick, 1990). The relationship between upper-echelon characteristics (i.e., TMT size and CEO dominance) and firm performance has been found to be significant in high-discretion environments but non-significant under conditions of low managerial discretion (Haleblian & Finkelstein, 1993).

In a further refinement of UET, scholars contend that UET predictions can be improved by paying closer attention to the structure of a firm’s TMT, including the extent of behavioural integration (Hambrick, 1995) and structural interdependence (Hambrick, Humphrey & Gupta, 2015). Many TMTs do not exhibit “team properties” but instead consist of semi-autonomous sub-teams of managers that engage in bilateral interactions with the CEO and have limited interactions with each other (Hambrick, 2007). Hambrick (1995) introduced the concept of “behavioural integration”, arguing that a TMT qualifies as a behavioural integrated team to the extent its members engage in mutual and collective interactions, such as information exchange, resource sharing, and joint decision-making. If TMTs are not behaviourally integrated, specific sub-groups of managers are responsible for certain types of decisions (Hambrick, 1995). For instance, a firm’s CEO, CFO, and top managers representing the functions of R&D and marketing (e.g., VP of R&D and VP of marketing) would be the relevant decision body in regard to R&D-related strategic decisions. It is, therefore, more plausible for upper echelons research to focus on the characteristics of those managers in charge of the specific decision under investigation rather than accounting for the TMT as a whole (UET2). Hambrick et al. (2015) suggested that a TMT is a meaningful entity only when there is high structural interdependence, which refers to the extent to which a TMT is structured in such a way that top managers have periodic and significant interactions with each other. When there is high structural interdependence, studying the characteristics of a firm’s TMT as a whole can yield strong predictions of strategic outcomes as top managers engage in joint decision-making. However, when managers operate independently and have distinct decision-making roles and responsibilities, there is no point in focusing on the TMT as the unit of analysis. In fact, it was empirically demonstrated that the association between TMT heterogeneity and organisational performance depends on the extent of TMT structural interdependence (Hambrick, Humphrey & Gupta, 2015).

The concept of “executive job demands” is thought to be another moderator of the upper echelons logic (Hambrick, Finkelstein & Mooney, 2005). Defined as “the degree to which a given executive experiences his or her job as difficult or challenging” (Hambrick, Finkelstein & Mooney, 2005: p473), executive job demands significantly influence the rationality of strategic decisions. Even though higher job demands may actually encourage executives to be more rational and comprehensive, the job pressures (e.g., lack of time, pressure to perform) urge managers to take mental shortcuts, rely on previous experiences, and inject a great deal of their dispositions when making decisions (Hambrick, 2007). As such, strategic choices closely reflect the characteristics of decision-makers when job demands are high. On the other hand, executives with lower job demands can afford to be more rational in their decision-making since they have the resources (e.g., time and attention) to thoroughly analyse strategic situations and search for solutions (Hambrick, Finkelstein & Mooney, 2005). Hence, their choices will closely match the objective conditions they confront.

Managerial power, or the extent to which managers have the requisite power/capacity to exert their will, is also thought to influence the extent to which top managers’ characteristics can influence strategic choices (Finkelstein, 1992). According to this line of thought, powerful managers have much more say in strategy discussions than others, and, consequently, their characteristics should be given more attention when predicting strategic choices. In fact, it was found that the relationship between TMT members’ characteristics and strategic outcomes is strengthened when considering the distribution of power among members of a firm’s TMT (Finkelstein, 1992).

Process mechanisms

Another stream of research has attempted to establish the underlying processes/mechanisms through which managerial characteristics shape strategic choices and resulting performance outcomes. According to the original UET model (Hambrick & Mason, 1984), observable managerial characteristics serve as proxy indicators of the psychological processes that affect strategic decision-making. However, this approach does not capture the black box processes that shape strategic outcomes (Carpenter, Geletkanycz & Sanders, 2004). Scholars have therefore suggested that the effect of upper echelon characteristics on strategic decisions and organisational performance is mediated through cognitive processes. These may include managerial attention (Cho & Hambrick, 2006), decision-making processes (e.g., decentralisation, communication, and comprehensiveness) (Papadakis & Barwise, 2002), as well as TMT-related processes such as conflict (Knight et al., 1999), psychological empowerment (Lin & Rababah, 2014), and group functioning (Peterson et al., 2003) (Figure 4).

For instance, Cho and Hambrick (2006) proposed “managerial attention” as a key mediator of the upper echelons logic because strategic action largely depends on what strategic stimuli managers direct their attention to. Managers with different characteristics differ in how they notice, interpret, and focus time and effort on strategic issues and action alternatives, thereby arriving at a different set of strategic decisions (Ocasio, 1997). Thus, it was empirically demonstrated that the relationship between TMT characteristics and strategy is partially mediated by managerial attention (Cho & Hambrick, 2006). Papadakis and Barwise (2002) suggested that upper-echelon characteristics influence strategic choices through their effect on four dimensions of the decision-making process: comprehensiveness/rationality, hierarchical decentralisation, lateral communication, and politicisation. It was reported that both CEO and TMT characteristics affect the process of decision-making, but in different ways. CEO characteristics influence the degree of hierarchical decentralisation, while characteristics of the TMT relate more to the dimensions of lateral communication and comprehensiveness.

In a different vein, scholarly work (Peterson et al., 2003) indicates that a CEO’s characteristics indirectly influence organisational outcomes by affecting the dynamics of the management team. Since CEOs have significant discretion over decisions about the composition and structure of their management teams, their personality characteristics could potentially shape the decision-making environment of the TMT. For example, CEOs high in “agreeableness” prefer cooperative and cohesive teams characterised by decentralised decision-making, whilst CEOs high in “conscientiousness” favour a centralised power structure. This, in turn, induces performance differences among firms. In confirmation of this logic, Peterson et al. (2003) empirically demonstrated that the relationship between CEO personality and organisational performance is fully mediated by TMT decision-making dynamics. Lin and Rababah (2014) proposed TMT psychological empowerment as another mediator of UET predictions. Defined as senior managers’ “collective beliefs in their autonomy and capability to perform meaningful work that can impact their organisation” (Lin & Rababah, 2014: p944), TMT psychological empowerment is strongly affected by characteristics of top-level managers, including CEO-TMT exchange quality and TMT personality composition. In turn, it was found that when executives feel empowered, they are more likely to arrive at strategic decisions of higher quality.

Applications

UET has been mainly applied in the field of management, but it has also sparked research across various other domains, including:  marketing (Chung & Low, 2022; Kashmiri & Mahajan, 2017), international business (Herrmann & Datta, 2005; Tihanyi et al., 2000), leadership (Waldman, Javidan & Varella, 2004; Lin & Rababah, 2014), psychology (Peterson et al., 2003; West & Anderson, 1996), accounting (Naranjo-Gil, Maas & Hartmann, 2009; Pavlatos, 2012) and economics (Bertrand & Schoar, 2003). Significant empirical support has been offered to the upper echelons logic, thereby highlighting its applicability across various disciplines and decision-making situations. Scholarly work has consistently documented managerial characteristics' influence on various strategic choices – such as strategic changes (Wiersema & Bantel, 1992; Waldman, Javidan & Varella, 2004), alliance formation (Eisenhardt & Schoonhoven, 1996), competitive attacks (Ferrier, 2001), international diversification (Tihanyi et al., 2000), innovation (Bantel & Jackson, 1989; West & Anderson, 1996), R&D investments (Kor, 2006), new product introductions (Kashmiri & Mahajan, 2017), marketing management (Chung & Low, 2022), and management accounting and control (Naranjo-Gil, Maas & Hartmann, 2009; Pavlatos, 2012) – and the resulting performance outcomes. Table 1 summarises the most commonly studied upper-echelon characteristics and strategic choice variables.

For instance, Chung and Low (2022) sought to understand the influence of CEO regulatory focus on myopic marketing management, which refers to the tendency to make short-term oriented marketing decisions. The authors showed that promotion-focused CEOs are more likely to engage in myopic marketing management as short-term performance aspirations drive their decisions. On the other hand, prevention-focused CEOs are less prone to making myopic marketing decisions, driven mainly by their need for security and loss avoidance. In line with UET predictions, Chung and Low (2022) find that strategic decisions (myopic marketing management) mediate the impact of upper echelon characteristics (CEO regulatory focus) on long-term organisational performance. Marketing scholars have also investigated whether and to what extent the inclusion of a chief marketing officer (CMO) in the top management team affects organisational performance (Germann, Ebbes & Grewal, 2015; Nath & Mahajan, 2008; Whitler, Krause & Lehmann, 2018). Considering that CMOs bring a customer perspective to the strategy table and facilitate decision-making (Bommaraju et al., 2019), it was found that firms benefit financially by employing CMOs on their management teams (Germann, Ebbes & Grewal, 2015). However, Nath and Mahajan (2008) reported neither a positive nor a negative effect of CMO presence on organisational performance.

Upper echelons research in the fields of accounting and economics has broadened the set of decision-makers and decision-making situations that are relevant to UET. For example, Pavlatos (2012) examined how chief financial officers’ (CFO) characteristics influence the use of cost-management systems for decision-making, control, and performance evaluation, while Naranjo-Gil et al. (2009) investigated the role of CFO characteristics in adopting management accounting innovations. It was found that firms with younger CFOs and CFOs with business-related educational backgrounds exhibit more comprehensive use of cost management systems (Pavlatos, 2012) and are more likely to adopt innovative management accounting systems (Naranjo-Gil, Maas & Hartmann, 2009).

In addition, scholars have established the applicability of UET beyond classic, demographic variables, and, therefore, have considered the effects on decision-making and the performance of senior managers’ political ideologies (Kashmiri & Mahajan, 2017), personality factors (Chatterjee & Hambrick, 2007), leadership behaviours (Waldman, Javidan & Varella, 2004), governance orientation (Kwee, Van Den Bosch & Volberda, 2011), power concentration (Greve & Mitsuhashi, 2007), network ties (Collins & Clark, 2003), and compensation (Luo, Wieseke & Homburg, 2012). As an example, Chatterjee et al. (2007) argued that narcissistic CEOs differ from non-narcissistic CEOs in how they make strategic decisions due to their inflated self-views and need for attention. The authors suggested that narcissistic CEOs are more likely to engage in novel and bold strategic actions with uncertain payoffs. In fact, CEO narcissism was found to be positively associated with strategic dynamism and intense acquisition activity. Firms led by narcissistic CEOs were also found to exhibit extreme (big wins or losses) and fluctuating performance. However, there were no significant performance differences between firms headed by narcissistic versus non-narcissistic chief executives. From a power relations perspective, Greve (2007) demonstrated that power concentration at the CEO level (i.e., when the CEO possesses excessive power compared to other organisational members) or TMT level (i.e., when a small number of senior managers have significantly more power than others) is associated with higher levels of strategic change. It was argued that power strongly affects the decision-making process as powerful managers favour decisions that signal and reinforce their position of power, such as strategic changes. Greve (2007:p.1200) pointed out that “strategic changes have a symbolic value because a high level of change indicates that the TMT has an active hand in strategy making”.

Scholars have established the applicability of UET across different national contexts, including both Western and Eastern countries (Geletkanycz & Black, 2001; Wiersema & Bantel, 1992; Wiersema & Bird, 1993), different industries, including both the services and manufacturing sectors (Lee & Park, 2006; Lin & Rababah, 2014; van Doorn, Heyden & Volberda, 2017), and different types of firms, including large, mature organisations as well as SMEs and newly founded companies (Carpenter, 2002; Escriba-Esteve, Sanchez-Peinado & Sanchez-Peinado, 2009; Reuber & Fischer, 1997). For instance, Geletkanycz and Black (2001) utilised data from 20 countries to confirm UET predictions that managerial characteristics (i.e., functional experience) exert significant influence on decision-making (i.e., the tendency to change organisational strategies). Lee and Park (2006) applied the upper echelons logic using data from 14 industries to find that firms headed by managers with heterogeneous characteristics are more likely to establish international alliances, which in turn leads to higher levels of internationalisation. Carpenter (2002) addressed the performance effects of TMT heterogeneity in large and medium-sized organisations, while Escriba-Esteve et al. (2009) established the link between managerial characteristics (e.g., age, education, previous experience), strategic behaviour and organisational performance in small and medium-sized enterprises.

Limitations

Three major limitations accompany the upper-echelon perspective. First, UET has been mainly criticised for a significant lack of attention to the underlying mechanisms through which top executives impact organisational outcomes – also known as the “black box problem” (Lawrence, 1997). The use of managers’ demographic and background characteristics as proxy indicators of their cognitive base and values places more emphasis on broad tendencies, thereby neglecting the actual psychological processes that drive strategic choices and performance (Neely et al., 2020). Although observable managerial characteristics are convenient to use and facilitate the reproducibility of empirical findings, they encapsulate more noise than pure psychological measures and are often imprecise and unreliable indicators of psychological variables (Markóczy, 1997; Priem, Lyon & Dess, 1999). Hence, ambiguous and inconsistent findings may be observed.

Another critique challenges the predictive power of UET (Carpenter, Geletkanycz & Sanders, 2004; Hoskisson et al., 2017; Hutzschenreuter, Kleindienst & Greger, 2012). The upper echelons' logic posits that managerial characteristics shape organisational outcomes, but desired organisational outcomes may influence the types of executives serving in a firm’s top management team (Finkelstein, Cannella & Hambrick, 2005). Top-level managers are often selected purposefully because they have the appropriate characteristics to execute actions desired by the board of directors (Hambrick, 2007). For instance, companies appoint new CEOs outside the organisation to overcome inertia and enact strategic change (Schepker et al., 2017). In this case, managers enact specific strategic actions due to a mandate rather than their personalised interpretation of strategic situations as posited by UET (Carpenter, Geletkanycz & Sanders, 2004). The relationship between managerial characteristics and organisational outcomes described by UET can be further confounded if we consider that strategic actions often hinge on a plethora of internal and/or external contingencies (Neely et al., 2020). For instance, firms that adopt a “prospector” strategy type typically pursue innovative strategic decisions not because of their executives’ volition but because of the ingrained character of a "prospector" strategy that encapsulates a constant search for new products and markets (Miles & Snow, 2003). In a similar vein, national institutions can place significant restrictions on executive actions. Scholars have demonstrated that managers have great latitude of action in countries with strong national values of individualism and tolerance for uncertainty (e.g., the US), but have little leeway to enact whatever actions they deem appropriate in countries where collectivism and uncertainty avoidance is high (e.g., Japan) (Crossland & Hambrick, 2011). Thus, the predictive strength of UET is minimal in the latter type of context. 

Finally, UET has been criticised for advocating the top management team as the unit of analysis (Carpenter, Geletkanycz & Sanders, 2004). This approach assumes that all senior managers contribute equally to decision-making, but the reality is that specific sub-groups of managers are primarily responsible for certain types of decisions (UET2). Furthermore, the factors (i.e., functional roles and group processes) that might influence each manager's contribution to group decision-making are not considered (Jensen & Zajac, 2004). Although strategic work is a shared activity, CEOs are the most powerful actors and are considered the principal architects of strategic decisions (Child, 1972). In fact, scholarly work has shown that CEO characteristics are significant predictors of strategic choices and organisational performance (Barker & Mueller, 2002; Bigley & Wiersema, 2002; Carpenter, Sanders & Gregersen, 2001; Chatterjee & Hambrick, 2007; Chung & Low, 2022). It has therefore been argued that the application of UET does not require a focus on TMTs as a whole (Hambrick, 2007; Hambrick, 2018).