Introduction
Why some chief executive officers (CEOs) pursue risk-taking at the firm level while others favor caution remains a foundational question in management (Arrfelt, Mannor, Nahrgang & Christensen, Reference Arrfelt, Mannor, Nahrgang and Christensen2018). Firm risk-taking reflects CEOs’ strategic decisions on critical actions and investments – such as R&D, acquisitions, or market entry – marked by uncertainty and the potential of high returns or significant losses (Arrfelt et al., Reference Arrfelt, Mannor, Nahrgang and Christensen2018; Hoskisson, Chirico, Zyung & Gambeta, Reference Hoskisson, Chirico, Zyung and Gambeta2017; Li, Liu & Dong, Reference Li, Liu and Dong2025). One research stream of firm risk-taking is CEO origin – whether an executive is promoted from within the firm as an insider or recruited from outside the firm as an outsider (Quigley, Hambrick, Misangyi & Rizzi, Reference Quigley, Hambrick, Misangyi and Rizzi2019). As this distinction has been widely debated in the literature on strategic decisions and firm performance (e.g., Chen & Hambrick, Reference Chen and Hambrick2012; Keil, Lavie & Pavicevic, Reference Keil, Lavie and Pavicevic2022; Schepker, Kim, Patel, Thatcher & Campion, Reference Schepker, Kim, Patel, Thatcher and Campion2017), outsider CEOs are more likely to engage in firm risk-taking than insiders (Zhang & Rajagopalan, Reference Zhang and Rajagopalan2010).
However, CEO origin alone cannot predict strategic decisions or firm performance reliably while some studies find the positive effect of outsider CEOs on risk-taking and others show greater caution (Bromiley & Rau, Reference Bromiley and Rau2016; Schepker et al., Reference Schepker, Kim, Patel, Thatcher and Campion2017). As Hambrick and Lee (Reference Hambrick and Lee2025) emphasize, besides, a CEO’s effectiveness depends not only on his or her origin but also on contingency conditions that shape his or her discretion and performance. While most studies pay attention to three dimensions of CEO origin – performance conditions (Keil et al., Reference Keil, Lavie and Pavicevic2022), contextual triggers (Liu & Xue, Reference Liu and Xue2020), and risk tendencies (Schepker et al., Reference Schepker, Kim, Patel, Thatcher and Campion2017) – from the firm-level perspective, they overlook the individual perspective to explore possible contingencies (Hambrick & Lee, Reference Hambrick and Lee2025).
There are two possible individual-level contingencies: CEOs’ temporal orientation and cognitive focus. For one thing, outsider CEOs are typically recruited and appointed with mandates for long-term transformation from the recruiting firm’s perspective. Despite that, the CEO’s temporal orientation of the environment, either short-termism or long-termism, influences how these mandates translate into risk-taking from the individual perspective (DesJardine & Shi, Reference DesJardine and Shi2021; Marginson & McAulay, Reference Marginson and McAulay2008). When temporal orientation emphasizes immediacy, for instance, CEOs who are short-termism would chase predictable gains at the expense of bold initiatives (Agnihotri, Bhattacharya & Satya Prasad, Reference Agnihotri, Bhattacharya and Satya Prasad2025; Kim & Nguyen, Reference Kim and Nguyen2024). By contrast, long-termism provides CEOs with more supportive conditions by legitimizing delayed returns, increasing tolerance for uncertainty, and encouraging exploratory commitments (Martin, Wiseman & Gomez-Mejia, Reference Martin, Wiseman and Gomez-Mejia2016; Nadkarni & Chen, Reference Nadkarni and Chen2014). Given the same origins, CEOs have different temporal orientations which affect risk-taking (DesJardine & Shi, Reference DesJardine and Shi2021; Martin et al., Reference Martin, Wiseman and Gomez-Mejia2016).
Another individual contingent factor is cognitive focus. Cognitive focus is a CEO’s cognitive framing of the external environment which influences his or her willingness to take risks by shaping how he or she recognizes and evaluates strategic opportunities (Arrfelt et al., Reference Arrfelt, Mannor, Nahrgang and Christensen2018; Eklund & Mannor, Reference Eklund and Mannor2021; Eklund, Raj & Eggers, Reference Eklund, Raj and Eggers2025). For example, attention breadth, one kind of cognitive focus, shapes how CEOs recognize opportunities and interpret uncertainty (Eklund & Mannor, Reference Eklund and Mannor2021). Specifically, narrow attention fosters CEOs to recognize opportunities through path dependence, while broad attention enhances them to enable bolder strategic choices through wide scanning and flexibility (McHugh & Duane, Reference McHugh and Duane2025). Given the same origins, CEOs also demonstrate distinct cognitive focus that influences risk-taking (Eklund & Mannor, Reference Eklund and Mannor2021; McHugh & Duane, Reference McHugh and Duane2025).
To address the research gap, we adopt the strategic perspective of microfoundations. This perspective emphasizes that firm-level outcomes such as firm risk-taking originate from microfoundations – individual characteristics, cognitions, and actions of decision-makers (Devinney, Reference Devinney2013; Felin, Foss & Ployhart, Reference Felin, Foss and Ployhart2015; Foss & Klein, Reference Foss and Klein2023). In this view, organizational outcomes are rooted not only in structural factors but also in executives’ behavioral patterns and cognitive orientations (Cristofaro, Augier, Lovallo, Abatecola & Leoni, Reference Cristofaro, Augier, Lovallo, Abatecola and Leoni2024; Durán & Aguado, Reference Durán and Aguado2022; Foss & Mazzelli, Reference Foss and Mazzelli2025; Ratten, Reference Ratten2025). Accordingly, we situate CEO origin, temporal orientation, and cognitive focus within a microfoundations framework to model how micro-level individual factors affect macro-level firm risk-taking. We argue that outsider CEOs are more likely to pursue higher levels of risk-taking, while this effect is constrained by short-termism, and reinforced by long-term orientations and broader attention breadth. By integrating these contingencies, we extend the insider–outsider comparison to capture both the classic contrasts and the microfoundational mechanisms that help reconcile prior inconsistencies in the literature. Testing the 20 years of panel data on the top 100 firms listed in the S&P 500 index, drawn from annual reports, CEO Letters to Shareholders (LTS), and financial statements, we found our premises supported.
The following section presents the theoretical background and hypothesis development. We then describe the methodology and data used to test our framework. The statistical and empirical results follow this. Finally, we provide our contributions, limitations, and avenues for future research.
Literature review and hypothesis
Microfoundations of firm risk-taking
Firm risk-taking refers to strategic decisions that involve uncertain outcomes made by top executives, particularly by CEOs and top management teams (Li et al., Reference Li, Liu and Dong2025). These decisions reflecting a willingness to commit resources despite uncertain returns including R&D investments, diversification moves, acquisitions, divestitures, and competitive actions (Hoskisson et al., Reference Hoskisson, Chirico, Zyung and Gambeta2017). Furthermore, firm risk-taking involves irreversible commitments and delayed feedback, making its ultimate performance implications difficult to evaluate in advance (Bromiley, Rau & Zhang, Reference Bromiley, Rau and Zhang2017; Xu, Zhou & Du, Reference Xu, Zhou and Du2019). Two essential features further characterize firm risk-taking: it demands sustained managerial effort (e.g. financial commitments or cognitive input) and entails outcomes that are not only uncertain but fundamentally unknowable at the time of decision (Weng & Kim, Reference Weng and Kim2023).
Traditional explanations of firm risk-taking emphasize both structural and incentive-based mechanisms (Hoskisson et al., Reference Hoskisson, Chirico, Zyung and Gambeta2017). In one side, the structural approach focuses on organizational and contextual conditions that shape strategic behavior. For example, He, Huang and Yang (Reference He, Huang and Yang2021) explore how the levels of firm aspiration affect risk-taking through the Behavioral Theory of the Firm. Meanwhile, Mueller-Saegebrecht (Reference Mueller-Saegebrecht2024) discusses how managers’ risk perceptions influence risky actions, drawing from Prospect Theory (Kahneman & Tversky, Reference Kahneman and Tversky1979). In another side, the incentive-based perspective emphasizes the alignment of managerial behaviors with shareholder interests. Nobre, Grable, Silva and Nobre (Reference Nobre, Grable, Silva and Nobre2018) examine how differing perceptions between principals and decision-makers shape risk-taking behavior, building on Agency Theory (Jensen & Meckling, Reference Jensen and Meckling1976). Benischke, Martin and Glaser (Reference Benischke, Martin and Glaser2019) further investigate the effect of CEO compensation structures on firm risk-taking using the Behavioral Agency Model (Wiseman & Gomez-Mejia, Reference Wiseman and Gomez-Mejia1998). Despite their valuable insights, these approaches often fall short in explaining why firms facing similar external conditions exhibit divergent strategic risk behaviors. Specifically, they tend to understate how individual differences in cognition and perception influence executive responses to uncertainty.
In response, the microfoundations perspective reorients attention to the individual-level factors, positing that firm-level phenomena originate from the cognitions, motivations, and behaviors of key decision-makers, including firm risk-taking (Felin & Foss, Reference Felin and Foss2005). Without linking macro-level constructs to individual actions, explanations lose clarity and miss the underlying causal logic (Abell, Felin & Foss, Reference Abell, Felin and Foss2008). Accordingly, this perspective emphasizes that organizational outcomes emerge from the aggregation of micro-level elements like individuals, their interactions, and the mechanisms through which their characteristics are expressed collectively (Cristofaro & Lovallo, Reference Cristofaro and Lovallo2022; Felin et al., Reference Felin, Foss and Ployhart2015). Crucially, this aggregation is not merely additive but often involves complex, emergent processes shaped by social interaction and contextual influence (Barney & Felin, Reference Barney and Felin2013). According to Foss and Lindenberg (Reference Foss and Lindenberg2013), sound microfoundations must account for the dynamic interplay between cognition and motivation, as these elements jointly shape how strategic goals are perceived, prioritized, and translated into action. The focus on individual-level variation is particularly relevant to firm risk-taking, as differences in managerial cognitive capabilities underpin how executives sense opportunities, evaluate strategic alternatives, and make high-stakes decisions under uncertainty (Durán & Aguado, Reference Durán and Aguado2022; Helfat & Peteraf, Reference Helfat and Peteraf2015; Mishra, Reference Mishra2023). Similarly, Heubeck and Meckl (Reference Heubeck and Meckl2023) highlight how dynamic CEO capabilities, particularly those rooted in individual cognition, social capital, and accumulated experience, constitute critical microfoundations of innovation and strategic renewal in uncertain settings. To better understand the underlying behaviors leading to differences in managerial decisions and actions, it is essential to examine how individual-level factors (e.g., behavioral and personality traits) shape strategic responses to risk-laden environments (Durán & Aguado, Reference Durán and Aguado2022; Ratten, Reference Ratten2025).
Recent developments in the microfoundations literature emphasize the importance of specifying the mechanisms through which individual-level traits influence organizational outcomes. For example, Palmié, Rüegger and Parida (Reference Palmié, Rüegger and Parida2023) highlight the value of multi-level approaches tracing how individual cognition and behavior interact with situational and organizational conditions. Moreover, Foss and Mazzelli (Reference Foss and Mazzelli2025) stress the role of interface mechanisms like hierarchical relationships and contextual constraints in shaping how executive cognition is enacted. Empirical research also supports the relevance of cross-level links. For instance, Bendig, Strese, Flatten, da Costa and Brettel (Reference Bendig, Strese, Flatten, da Costa and Brettel2018) show that CEO personality traits affect firm capabilities indirectly through internal knowledge structures and organizational capital. Taken together, these insights reinforce the need to account for microfoundations including both individual dispositions and contextual moderators when explaining strategic behaviors.
CEO origin and firm risk-taking
CEO origin constitutes one of critical microfoundations that shapes how firms respond to uncertainty and initiate strategic action (Felin et al., Reference Felin, Foss and Ployhart2015; Hambrick & Lee, Reference Hambrick and Lee2025; Quigley et al., Reference Quigley, Hambrick, Misangyi and Rizzi2019). Moving beyond demographic classification, recent scholarship conceptualizes CEO origin as a proxy for variation in executives’ cognitive schemas, experiential repertoires, and embedded social capital (Haque, Choi, Lee & Wright, Reference Haque, Choi, Lee and Wright2022; Kunisch, Menz & Cannella, Reference Kunisch, Menz and Cannella2019). These differences inform how CEOs attend to, interpret, and act upon complex strategic issues including those involving risk. CEO origin is generally classified into two categories: insiders, who are promoted from within the firm, and outsiders, who are hired from external organizations. While both types of CEOs can impact firm strategy in meaningful ways, they differ systematically in terms of performance conditions, contextual triggers, and risk tendencies. A summary of these differences is presented in Table 1.
Comparison of insider vs. outsider CEOs
Insider CEOs are typically appointed when boards seek continuity and cohesion, especially under stable performance conditions (Quigley et al., Reference Quigley, Hambrick, Misangyi and Rizzi2019). Their deep embeddedness in firm-specific routines, culture, and social capital provides reliable knowledge for managing resources and anticipating implementation challenges (Castanias & Helfat, Reference Castanias and Helfat1991; Schepker et al., Reference Schepker, Kim, Patel, Thatcher and Campion2017). This embeddedness, however, also reinforces incremental adaptation and preservation of the status quo, making insiders more cautious and risk-averse in their strategic choices (Chen & Hambrick, Reference Chen and Hambrick2012; Chiu, Johnson, Hoskisson & Pathak, Reference Chiu, Johnson, Hoskisson and Pathak2016; Zhang & Rajagopalan, Reference Zhang and Rajagopalan2010).
In contrast, outsider CEOs are frequently recruited when boards call for transformation and renewal, especially in situations marked by performance decline, stagnation, or discredited leadership (Quigley et al., Reference Quigley, Hambrick, Misangyi and Rizzi2019). Outsiders bring diverse experiences and fresh viewpoints that empower them to challenge outdated practices and expand strategic possibilities (Haque et al., Reference Haque, Choi, Lee and Wright2022; Liu & Xue, Reference Liu and Xue2020). With fewer constraints imposed by internal routines and politics, they are granted broader discretion to enact bold organizational changes (Chiu et al., Reference Chiu, Johnson, Hoskisson and Pathak2016). Outsiders are thus widely expected to pursue riskier initiatives and act as catalysts for renewal. For one thing, they can drive strategic change and innovation; for another, they may incur short-term disruption, steep learning curves, and integration challenges that sometimes lead to long-term underperformance (Hambrick & Lee, Reference Hambrick and Lee2025; Hughes, Hughes, Mellahi & Guermat, Reference Hughes, Hughes, Mellahi and Guermat2010; Schepker et al., Reference Schepker, Kim, Patel, Thatcher and Campion2017).
Accordingly, we argue that outsider CEOs are more likely to engage in firm risk-taking due to performance conditions, contextual triggers, and risk tendencies that differentiate them from insiders. Firstly, outsider CEOs possess interpretive schemas shaped by diverse external experiences, enabling them to question legacy practices and identify unconventional opportunities under uncertainty (Firk, Hennig, Meier & Wolff, Reference Firk, Hennig, Meier and Wolff2024; Helfat & Peteraf, Reference Helfat and Peteraf2015). Secondly, their lack of embeddedness in internal networks reduces sensitivity to firm norms and politics, granting them greater autonomy to deviate from the status quo and enact bold change (Schepker et al., Reference Schepker, Kim, Patel, Thatcher and Campion2017). Thirdly, outsider CEOs are frequently appointed during periods of poor performance or strategic drift, where they are granted broader discretion to pursue aggressive repositioning strategies (Quigley et al., Reference Quigley, Hambrick, Misangyi and Rizzi2019). In contrast, insider CEOs tend to uphold continuity through incremental adjustments, shaped by internal social capital and firm-specific knowledge (Zhang & Rajagopalan, Reference Zhang and Rajagopalan2010). As these comparisons suggest that outsider CEO origin fosters a stronger propensity for firm risk-taking, we establish the first hypothesis:
Hypothesis 1. Outsider CEOs are more likely to engage in firm risk-taking than insider CEOs.
Contextual contingencies
While CEO origin provides an entry point into understanding executive influence on firm risk-taking, its effects are rarely uniform across organizations (Bromiley & Rau, Reference Bromiley and Rau2016; Kunisch et al., Reference Kunisch, Menz and Cannella2019). Outsider CEOs, in particular, do not act in a vacuum – their strategic behaviors are conditioned by how internal environments interact with their cognitive framing and strategic intent (Hambrick & Lee, Reference Hambrick and Lee2025; Quigley et al., Reference Quigley, Hambrick, Misangyi and Rizzi2019). Within the microfoundations perspective, individual-level traits shape firm outcomes through contextual and organizational mechanisms that channel, enable, or constrain the expression of executive cognition and behavior (Arndt, Galvin, Jansen, Lucas & Su, Reference Arndt, Galvin, Jansen, Lucas and Su2022; Felin et al., Reference Felin, Foss and Ployhart2015; Foss & Mazzelli, Reference Foss and Mazzelli2025). These contextual mechanisms help explain the inconsistent effects of outsider CEOs reported in prior studies and underscore the importance of factors such as temporal orientation and cognitive focus, which could moderate how executive intent is translated into firm-level outcomes (Bromiley & Rau, Reference Bromiley and Rau2016; Palmié et al., Reference Palmié, Rüegger and Parida2023).
Temporal orientation
Temporal orientation represents a critical contextual element within the microfoundations perspective, shaping how decision-makers allocate attention and evaluate strategic options over time (Cristofaro et al., Reference Cristofaro, Augier, Lovallo, Abatecola and Leoni2024; Laverty, Reference Laverty1996). It defines the time horizon CEOs emphasize when assessing risk, ranging from a focus on short-term performance metrics to a long-term orientation centered on future strategic outcomes (DesJardine & Shi, Reference DesJardine and Shi2021; Nadkarni & Chen, Reference Nadkarni and Chen2014). Recent findings suggest that a CEO’s temporal orientation moderates how executive traits shape strategic behavior (Agnihotri et al., Reference Agnihotri, Bhattacharya and Satya Prasad2025), thereby shaping whether bold, high-variance actions are supported or constrained.
Within this framework, short-termism denotes a cognitive and organizational preference for near-term, tangible outcomes, often reinforced by internal accountability systems and external capital market demands (Brochet, Loumioti & Serafeim, Reference Brochet, Loumioti and Serafeim2015; Marginson & McAulay, Reference Marginson and McAulay2008). By contrast, long-termism supports openness to delayed rewards and uncertain investments, favoring innovation and strategic renewal (Martin et al., Reference Martin, Wiseman and Gomez-Mejia2016; Nadkarni & Chen, Reference Nadkarni and Chen2014). When CEOs are exposed to short-termist pressures, they tend to underinvest in high-risk initiatives such as R&D, acquisitions, or exploratory growth (Bushee, Reference Bushee1998; Kleinknecht, Haq, Muller & Kraan, Reference Kleinknecht, Haq, Muller and Kraan2020). This is consistent with recent research emphasizing that the impact of executives is temporally contingent, unfolding differently across short- versus long-term time horizons (Engelen, Brettel & Heubeck, Reference Engelen, Brettel and Heubeck2025).
Short-termism imposes behavioral constraints by narrowing the range of acceptable strategic options, amplifying aversion to uncertainty, and reducing the likelihood of long-term commitments (Gryglewicz, Mayer & Morellec, Reference Gryglewicz, Mayer and Morellec2020; Kim & Nguyen, Reference Kim and Nguyen2024; Smulowitz, Cossin & Lu, Reference Smulowitz, Cossin and Lu2023). For instance, Kim and Nguyen (Reference Kim and Nguyen2024) indicate that some managers may prioritize immediate financial gains at the cost of sustainable development, whereas other managers who adopt a more patient perspective might strategically position their firms for long-term success. CEOs embedded in such environments often deprioritize uncertain ventures due to concerns over delayed feedback, performance evaluations, and reputational risk (DesJardine & Shi, Reference DesJardine and Shi2021; Martin et al., Reference Martin, Wiseman and Gomez-Mejia2016). These dynamics discourage firm-level risk-taking and inhibit the pursuit of bold, future-oriented actions.
These constraints are especially problematic for outsider CEOs. Unlike insiders, they lack firm-specific knowledge, established internal legitimacy, and trust-based relationships with key stakeholders (Keil et al., Reference Keil, Lavie and Pavicevic2022; Zhang & Rajagopalan, Reference Zhang and Rajagopalan2010). This liability of outsidership (Shen & Cannella, Reference Shen and Cannella2002) heightens board and investor scrutiny, as stakeholders rely more heavily on observable short-term results to evaluate effectiveness, leaving outsider CEOs more exposed to performance monitoring and stakeholder skepticism (Hambrick & Lee, Reference Hambrick and Lee2025; Keil et al., Reference Keil, Lavie and Pavicevic2022). Under conditions of short-termism, this heightened scrutiny diminishes their discretion to enact high-variance strategies. Even though CEOs appointed with a mandate for transformation may revert to safer, performance-driven behavior when temporal pressures are high (Rau & Bromiley, Reference Rau and Bromiley2025), short-termism could moderate the relationship between CEO origin and firm risk-taking.
On the other hand, long-termism creates more supportive conditions for outsider CEOs to pursue risk-taking. A long-term orientation broadens the time horizon over which performance is evaluated, reducing immediate pressures for predictable gains and enhancing tolerance for delayed outcomes (Martin et al., Reference Martin, Wiseman and Gomez-Mejia2016; Nadkarni & Chen, Reference Nadkarni and Chen2014). This orientation legitimizes exploratory investments such as R&D and strategic renewal, thereby aligning with outsider CEOs’ mandate to pursue transformation (Quigley et al., Reference Quigley, Hambrick, Misangyi and Rizzi2019; Zhang & Rajagopalan, Reference Zhang and Rajagopalan2010). With fewer constraints from near-term accountability, outsider CEOs under long-termist conditions can more freely enact bold strategies, take calculated risks, and reposition firms for future opportunities. Therefore, we establish the following hypotheses:
Hypothesis 2a. Short-termism weakens the positive impact of outsider CEOs on firm risk-taking.
Hypothesis 2b. Long-termism strengthens the positive impact of outsider CEOs on firm risk-taking.
Cognitive focus
Cognitive focus represents another contextual dimension in the microfoundations perspective, shaping how executives filter, interpret, and respond to complex strategic stimuli (Rau & Bromiley, Reference Rau and Bromiley2025). Rather than assuming uniform executive decision-making, the microfoundations lens emphasizes bounded rationality and selective attention as key mechanisms through which individual-level traits influence firm-level outcomes (Durán & Aguado, Reference Durán and Aguado2022; Felin & Foss, Reference Felin and Foss2005; Helfat & Peteraf, Reference Helfat and Peteraf2015). Attention governs where CEOs allocate limited cognitive resources and thus determines which strategic issues are recognized and prioritized under conditions of uncertainty (Gavetti, Reference Gavetti2005; Ocasio, Reference Ocasio1997).
Among the many constructs that capture executive cognition, attention breadth is uniquely positioned to explain firm risk-taking because it emphasizes the scope of strategic issues recognized rather than the intensity of focus devoted to a single domain. Alternative constructs such as absorptive capacity or environmental scanning highlight the depth or amount of information processing, but they do not capture whether executives are cognitively oriented toward a narrow or broad range of domains (Brielmaier & Friesl, Reference Brielmaier and Friesl2023; Ocasio, Reference Ocasio2011). Attention breadth directly determines the diversity of opportunities that enter executives’ consideration set, expanding the pool of strategic alternatives available under uncertainty (Eklund & Mannor, Reference Eklund and Mannor2021; Posen, Keil, Kim & Meissner, Reference Posen, Keil, Kim and Meissner2018). This scope-oriented mechanism provides a distinct rationale for why attention breadth is central for linking CEO origin to risk-taking.
Attention breadth, therefore, captures the extent to which CEOs engage with a wide array of strategic issues. Defined as the diversity of domains that occupy executive focus, it reflects the inclusiveness and range of strategic scanning (Eklund & Mannor, Reference Eklund and Mannor2021). A broader attentional scope integrates opportunity recognition with richer interpretation of environmental cues and more flexible cognitive framing, together enabling CEOs to navigate complex, high-stakes decisions such as risk-taking (Eggers & Kaplan, Reference Eggers and Kaplan2009; Li, Maggitti, Smith, Tesluk & Katila, Reference Li, Maggitti, Smith, Tesluk and Katila2013). Empirical studies suggest that attention breadth enhances innovation and exploration by equipping decision-makers with a more expansive mental map of the competitive landscape (Mack, Cho & Yi, Reference Mack, Cho and Yi2024; Posen et al., Reference Posen, Keil, Kim and Meissner2018).
By scanning a more diverse set of external and internal cues, CEOs with broader attentional scope are better positioned to recognize novel opportunities, challenge entrenched assumptions, and pursue exploratory initiatives such as long-term investments and strategic experimentation (Brielmaier & Friesl, Reference Brielmaier and Friesl2023; Eklund & Mannor, Reference Eklund and Mannor2021; Pan, McNamara, Devers & Yonish, Reference Pan, McNamara, Devers and Yonish2025). Such breadth enables the construction of a dynamic and evolving strategic agenda that channels cognitive resources toward future-oriented initiatives (Ocasio & Joseph, Reference Ocasio and Joseph2018). Within a microfoundations framework, attention breadth serves as a behavioral mechanism through which individual-level variation in information processing is translated into firm-level actions (Cristofaro et al., Reference Cristofaro, Augier, Lovallo, Abatecola and Leoni2024; Foss & Mazzelli, Reference Foss and Mazzelli2025).
For outsider CEOs, who often face the dual challenges of limited firm-specific knowledge and heightened expectations for transformation, attention breadth serves as a crucial enabling mechanism that helps translate their external perspectives into bold organizational change (Brielmaier & Friesl, Reference Brielmaier and Friesl2023; Eklund & Mannor, Reference Eklund and Mannor2021; Quigley et al., Reference Quigley, Hambrick, Misangyi and Rizzi2019). A broader attention helps compensate for these liabilities by facilitating rapid sensemaking, expanding cognitive search, and supporting novel strategic choices (Firk et al., Reference Firk, Hennig, Meier and Wolff2024). In this sense, the breadth of attention mitigates outsider CEOs’ informational disadvantage and empowers them to pursue high-variance initiatives. It then amplifies outsider CEOs’ capacity to enact strategic change characterized by high risk, thereby reinforcing their propensity for risk-taking in ambiguous environments. Accordingly, we establish the third hypothesis:
Hypothesis 3. Attention breadth strengthens the positive impact of outsider CEOs on firm risk-taking.
Figure 1 presents the conceptual framework that illustrates these relationships and the moderating roles of temporal and cognitive contexts.
Conceptual framework.
Methodology
Sample and data
We test our hypotheses by utilizing a panel dataset comprising the top 100 firms listed in the S&P 500 index. These firms were selected due to their significant size, prominence, and well-established governance structures, making them suitable for examining how CEO origin interacts with temporal and cognitive contexts to shape firm risk-taking. The initial list was based on consistent representation and market capitalization over the sample period. The chosen period, 2000–2020, was selected to ensure sufficient historical data was available and able to capture long-term risk-taking and leadership change in a sampling firm. This timeframe allows for an analysis that spans various market conditions and regulatory shifts, offering a comprehensive perspective on risk-taking across different environments. After accounting for data availability and missing information, the final dataset comprises 100 firms, with 1,992 firm-year observations for financial variables and 1,673 for moderator variables. The reduction of observation reflects incomplete LTS coverage or firms founded in later years.
We integrate data from multiple archival sources to capture both financial indicators and managerial cognitive attributes. First, CEO demographic and succession data were manually extracted from annual reports and proxy statements. Firm-level financial data were obtained from firms’ financial statements. Control variables such as financial, operational, and market valuation metrics were derived to account for firm-level heterogeneity. Second, we constructed measures of CEO temporal orientation (short-termism vs. long-termism) and attention breadth using textual data from Letters to Shareholders (LTS), a section of annual reports widely recognized for conveying managerial priorities and strategic framing (Cho & Hambrick, Reference Cho and Hambrick2006; Eggers & Kaplan, Reference Eggers and Kaplan2009; Li et al., Reference Li, Maggitti, Smith, Tesluk and Katila2013; Nadkarni & Chen, Reference Nadkarni and Chen2014). These documents were collected from various source including annual reports and proxy statements, the investor relations sections of the firms’ websites and the U.S. Securities and Exchange Commission’s EDGAR database. Details of the textual analysis procedures and dictionary-based measures are provided in the following measurement section.
Variables and measurement
Dependent variable
Risk-Taking. To capture firm-level strategic risk-taking, we adopt a composite measure aligned with prior research in behavioral agency theory and strategic decision-making (e.g., Benischke et al., Reference Benischke, Martin and Glaser2019; Kish-Gephart & Campbell, Reference Kish-Gephart and Campbell2015; Martin, Gomez-Mejia & Wiseman, Reference Martin, Gomez-Mejia and Wiseman2013). Following Resick et al. (Reference Resick, Nadkarni, Chu, Chen, Lien, Margolis and Shao2023), we operationalize risk-taking as a firm’s financial commitment to three categories of high-risk, future-oriented investments: long-term debt (financing), R&D expenditures (innovation), and capital expenditures (asset commitments). Each represents an irreversible allocation of resources under uncertainty, and together they provide a multidimensional view of risk-taking behavior (Miller & Bromiley, Reference Miller and Bromiley1990). Each component was log-transformed to correct for skewness and enhance interpretability before aggregation to form a composite risk-taking score for each firm-year. The three indicators exhibit comparable distributions, with mean values of 6.88 (SD = 1.55) for R&D expenditures, 6.41 (SD = 2.18) for capital expenditures, and 8.56 (SD = 2.21) for long-term debt. We validated the aggregation of these components into a composite index by conducting a Cronbach’s alpha test, which yielded a reliability coefficient of 0.627. This level is acceptable given the conceptual heterogeneity of risk-taking domains and exceeds the commonly accepted 0.6 threshold (Hair, Black, Babin & Anderson, Reference Hair, Black, Babin and Anderson2010). The three log-transformed components are also moderately correlated (r = 0.48–0.61), indicating sufficient shared variance while retaining distinct information. Accordingly, the composite provides a more reliable and comprehensive proxy than any single indicator, which could overweight one channel of risk-taking. This approach provides a theoretically grounded and empirically validated proxy for CEO-driven risk-taking behavior (Benischke et al., Reference Benischke, Martin and Glaser2019; Martin et al., Reference Martin, Gomez-Mejia and Wiseman2013). To account for temporal dynamics and smooth volatility, we compute a 3-year rolling average of the composite score. A higher score indicates a greater commitment to resource allocation in areas associated with elevated strategic risk. Because this variable is constructed as a 3-year rolling average (t, t–1, t–2), pre-sample financial data from t–1, t–2 were collected to avoid losing early firm-years. Variation in N occurs only when firms lacked complete financial records in these years or firms founded later.
Independent variable
CEO Origin. CEO origin captures whether the chief executive officer was promoted from within the firm or hired externally. Consistent with prior work on CEO succession and strategic behavior (e.g., Quigley et al., Reference Quigley, Hambrick, Misangyi and Rizzi2019; Zhang & Rajagopalan, Reference Zhang and Rajagopalan2010), we classify CEOs as outsiders if they joined the firm fewer than two years before assuming the CEO role, and as insiders if they had been with the firm for at least two years prior to promotion. This distinction reflects differences in executives’ familiarity with internal routines, embeddedness in organizational structures, and cognitive framing of strategic priorities. We code CEO origin as a binary variable: 1 for outsider CEOs and 0 for insiders. Across the sample period (2000–2020), this classification identifies 269 outsider appointments (13.5% of firm-year observations). Outsider appointments within rolling windows are retained, as both the dependent variable (3-year rolling) and moderators (5-year rolling) capture firm-level conditions inherited by new CEOs. To ensure that observed effects reflect CEO origin rather than general succession shocks, we include a control for CEO change in all models.
Moderator variables
Temporal Orientation. To capture temporal orientation, we conducted an automated text analysis based on the linguistic content of LTS disclosed in annual reports. Following Brochet et al. (Reference Brochet, Loumioti and Serafeim2015), we operationalize time orientation using two complementary measures: short-termism and long-termism. The short-term dictionary includes 16 immediacy-related terms such as ‘day,’ ‘days,’ ‘daily,’ ‘short-run,’ and ‘short-term,’ while the long-term dictionary consists of 15 extended-horizon terms such as ‘year,’ ‘years,’ ‘yearly,’ ‘long-run,’ and ‘long-term’ (Brochet et al., Reference Brochet, Loumioti and Serafeim2015; Li, Reference Li2010). The full lists of terms are provided in Appendix A1. These dictionaries have been widely validated in research on time orientation and disclosure framing (Flammer & Bansal, Reference Flammer and Bansal2017; Gamache & McNamara, Reference Gamache and McNamara2019; Smulowitz et al., Reference Smulowitz, Cossin and Lu2023).
For each firm-year observation, we parsed the LTS into a term-document matrix, tokenized the text, and counted the occurrences of short-term and long-term words (Smulowitz et al., Reference Smulowitz, Cossin and Lu2023). On average, LTS contained about 123 short-term words and 65 long-term words out of roughly 1,538 total words. This distribution indicates that while both temporal orientations are represented in CEO discourse, short-term references are more frequent, consistent with prior findings on disclosure framing.
To capture temporal patterns and smooth year-to-year fluctuations, we first computed 5-year rolling averages for both measures, using pre-sample data (t, t–1 … t–4) to construct valid scores from 2000 onward. Next, to ensure that each measure reflects its distinct emphasis, we constructed residualized scores by regressing short-term counts on long-term counts and vice versa, following Brochet et al. (Reference Brochet, Loumioti and Serafeim2015). Specifically, we obtained the residualized values using the following equations:
\begin{equation}{\text{ShortTerm\_resi}}{{\text{d}}_i} = {\text{ShortTer}}{{\text{m}}_i} - {\hat \beta _0} - {\hat \beta _1}\left( {{\text{LongTer}}{{\text{m}}_i}} \right)\end{equation}
\begin{equation}{\text{LongTerm\_resi}}{{\text{d}}_i} = {\text{LongTer}}{{\text{m}}_i} - {\hat \alpha _0} - {\hat \alpha _1}\left( {{\text{ShortTer}}{{\text{m}}_i}} \right)\end{equation}These residuals represent the unique variance of each construct after removing their shared component. Variation in N reflects firms without LTS coverage in earlier years or those founded later.
Attention Breadth. We measure CEO strategic attention breadth using a dictionary-based text analysis of CEO LTS from 2000 to 2020, following prior work (Eklund & Mannor, Reference Eklund and Mannor2021; Pan et al., Reference Pan, McNamara, Devers and Yonish2025). This approach is widely validated in the strategy literature as a reliable way to capture managerial attention to specific concepts where prior expectations exist (e.g., Dutt & Joseph, Reference Dutt and Joseph2019; Eggers & Kaplan, Reference Eggers and Kaplan2009; Eklund et al., Reference Eklund, Raj and Eggers2025). Compared with inductive text methods such as topic modeling, dictionary-based analysis provides a transparent link between theoretical constructs and the language used by managers, making it suited for measuring attention breadth. The 13-category strategic issue dictionary developed by Eklund and Mannor (Reference Eklund and Mannor2021) serves as the foundation. It covers domains such as alliance partner strategies; customer oriented; external stakeholder management; financial and risk management; internal organization; low-cost and efficiency; mergers and acquisitions firm scope; new market entry; resource and capability development; social strategies; product and innovation; and business model innovation. To illustrate, the dictionary includes terms like alliance and joint-venture (alliance partner strategies), customer focus and customer value (customer orientation), government and regulation (external stakeholder management), and capital and debt (financial and risk management). The complete dictionaries for all 13 categories are provided in Appendix A2.
These categories were inductively derived from three practitioner and academic sources and validated by management scholars in the U.S. and Europe. Together, the dictionaries cover about 300 terms, with 10–40 words per category (Pan et al., Reference Pan, McNamara, Devers and Yonish2025), providing broad yet balanced coverage of issues commonly addressed in CEO communications. On average, 108 words per letter were classified into one of the 13 strategic categories. The most frequently coded domains were financial and risk management strategies, customer-oriented strategies, and product/innovation strategies, while less frequent categories included mergers and acquisitions/firm scope, performance management, and business model innovation. This distribution indicates that the dictionary captures a meaningful and balanced subset of CEO discourse devoted to strategic issues, consistent with prior validations (Pan et al., Reference Pan, McNamara, Devers and Yonish2025). While some words in the texts fall outside these categories, they are primarily generic or non-strategic (e.g., greetings, narrative connectors, formalities) rather than systematic domains of strategic content. Thus, the 13 categories provide comprehensive coverage of the strategic issue space typically addressed in CEO discourse and are well suited to capture variation in attention breadth.
Following Eklund and Mannor’s (Reference Eklund and Mannor2021) procedure, we parsed the text into a term-document matrix, tokenized the content, and counted the frequency of terms corresponding to each strategic issue category. For each firm-year observation, we then converted raw word counts into proportions by dividing each category’s word count by the total number of categorized strategic words in the respective CEO letter. This normalization ensures that attention breadth reflects the distribution of attention across domains rather than the absolute volume of disclosure, which can vary widely across firms and years due to differences in report length or writing style. By focusing on proportions, the measure captures how evenly a CEO allocates attention across categories, independent of overall verbosity. In doing so, it avoids confounding breadth with letter length, preventing firms with unusually long reports from appearing to have broader attention simply because they use more words. Rather than diluting acute effects, this procedure isolates relative allocation consistent with Eklund and Mannor’s (Reference Eklund and Mannor2021) validated approach.
Using these proportions, we computed a Herfindahl index of concentration and subtracted it from one to obtain a breadth score. For firm i at time t, attention breadth (
$a{b_{i,t}}$) was calculated using the following modified Herfindahl formula, where the proportion of words linked to each strategic issue (
${p_{i,k,t}}$) is defined below:
\begin{equation}a{b_{i,t}} = 1 - \frac{{\sum\limits_{k = 1}^{13}{{\left( {{p_{i,k,t}}} \right)}^2}}}{{{{\left( {\sum\limits _{k = 1}^{13}{p_{i,k,t}}} \right)}^2}}}\end{equation}
\begin{equation}{p_{i,k,t}} = \frac{{300}}{{13 \times {n_k}}} \times \frac{{{c_{i,k,t}}}}{{{q_{i,t}}}}\end{equation} where 
${c_{i,k,t}}$ is the count of words linked to strategic issue k, 
${q_{i,t}}$ is the total number of words in the letter, and 
${n_k}$ denotes the number of terms in the respective dictionary. The Herfindahl-based approach is critical because it differentiates managers who spread attention evenly across many domains from those who focus primarily on a few domains while superficially mentioning others (Eklund et al., Reference Eklund, Raj and Eggers2025). Similar dispersion-based measures are widely used in related literatures, such as revenue diversification (Palepu, Reference Palepu1985) and technological knowledge breadth (Eklund, Reference Eklund2022). To capture sustained attentional scope and smooth annual fluctuations, we computed 5-year rolling averages, using pre-sample data (t, t–1 … t–4) to construct valid scores from 2000 onward. Variation in N reflects firms without LTS coverage in these earlier years or firms founded later.
Control variables
To mitigate alternative explanations and isolate the effects of CEO origin and behavioral moderators on firm risk-taking, we included a comprehensive set of control variables at the individual, team, and firm levels. At the individual level, we controlled for CEO experience, measured as a dummy variable coded 1 if the executive had previously served as CEO in another firm and 0 otherwise, as such experience may shape cognitive style, strategic preferences, and risk propensity (Harrison, Boivie, Stern & Porac, Reference Harrison, Boivie, Stern and Porac2024; Li et al., Reference Li, Liu and Dong2025). At the management level, we accounted for top management team (TMT) size and average tenure, as TMT composition affects collective decision-making and strategic behavior (Bendig et al., Reference Bendig, Strese, Flatten, da Costa and Brettel2018; He et al., Reference He, Huang and Yang2021; Lu, Zhou, Luan & Deng, Reference Lu, Zhou, Luan and Deng2024). We also controlled for CEO change, a dummy coded 1 if a leadership transition occurred in a given year, to disentangle the effect of CEO origin from the broader consequences of succession events (Chiu et al., Reference Chiu, Johnson, Hoskisson and Pathak2016). At the firm level, we controlled for firm age (in years since founding), which captures structural inertia and the tendency toward conservative strategies (Back, Rosing, Dickler, Kraft & Bausch, Reference Back, Rosing, Dickler, Kraft and Bausch2020; Bendig et al., Reference Bendig, Strese, Flatten, da Costa and Brettel2018). We also included R&D intensity (R&D expenditures/ total assets), slack resources (long-term debt/total assets), asset turnover (operating income/total assets), and operational capacity (gross profit), which together capture financial and operational factors relevant to risk-taking (Eklund & Mannor, Reference Eklund and Mannor2021; Lee, Reference Lee2023). To control for firm performance expectations, we also included Tobin’s Q as a proxy for market-based valuation and investment attractiveness (Benischke et al., Reference Benischke, Martin and Glaser2019). Finally, we included year fixed effects through year dummy variables to control for temporal shocks, macroeconomic cycles, and industry-level trends that could impact firm risk-taking (Benischke et al., Reference Benischke, Martin and Glaser2019).
Estimation
To determine the appropriate panel estimation method, we conducted a Hausman test comparing fixed and random effects specifications. The test rejected the null hypothesis that differences in coefficients are not systematic (χ2(10) = 97.67, p < 0.001), indicating that the random effects estimator is inconsistent. Thus, all the main results that we describe include firm fixed effects. This method is well-suited for longitudinal data as it accounts for unobserved, time-invariant firm-level heterogeneity that could bias estimates. Unlike pooled OLS, fixed-effects models isolate within-firm variation over time, reducing omitted variable bias and addressing endogeneity concerns from stable organizational factors such as corporate culture or governance structures (Wooldridge, Reference Wooldridge2011). Given that our key independent and moderating variables – CEO origin, temporal orientation, and attention breadth – vary at the firm-year level, this approach allows us to estimate their effects more precisely. To examine moderation, we constructed interaction terms between CEO origin and each behavioral variable. A statistically significant interaction term would indicate that the relationship between outsider CEO origin and firm risk-taking depends on the firm’s temporal orientation or the CEO’s cognitive focus.
Results
Table 2 reports descriptive statistics and pairwise correlations. The average short-termism score is 4.49 (SD = 0.804), indicating that while short-term language appears regularly in CEO discourse, its emphasis varies considerably across firms and years. Long-termism averages 3.84 (SD = 0.733), showing that forward-looking references are somewhat less frequent but still present. Attention breadth averages 0.849 (SD = 0.036), suggesting CEOs generally maintain a broad strategic focus, though variation exists in how evenly attention is distributed. On average, TMTs consist of 10.77 members with 5.34 years of tenure, and firms are about 73 years old.
Descriptive statistics and correlations
Note: Correlations with absolute value greater than .05 are significant at p < .05 (two-tailed tests).
In the correlation matrix, risk-taking is positively and significantly associated with attention breadth (r = 0.145, p < .05), operational capacity (gross profit; r = 0.769, p < .05), and TMT size (r = 0.177, p < .05). By contrast, it is negatively associated with R&D intensity (r = –0.197, p < .05), asset turnover (r = –0.223, p < .05), and Tobin’s Q (r = –0.121, p < .05). Short-termism is significantly and positively correlated with risk-taking (r = .168, p < .05), as is long-termism (r = .340, p < .05). To assess potential multicollinearity, we examined both the pairwise correlation matrix and variance inflation factors (VIFs). No correlation exceeded 0.85, and the mean VIF values were: Model 1 = 1.85, Model 2 = 1.83, Model 3 = 1.82, Model 4 = 1.81, and Model 5 = 2.18. All values fall well below conventional thresholds of concern (Hair et al., Reference Hair, Black, Babin and Anderson2010), suggesting that multicollinearity is not a concern in our models.
Table 3 reports the results from the fixed-effects regression analyses. The number of observations varies across models because the moderator variables are derived from LTS. To construct valid rolling averages, we use pre-sample data from t–4; variation in N therefore reflects firms without sufficient LTS coverage in these earlier years or firms founded later. Accordingly, all the models cover around 1,673 to 1,992 firm-years observations across 100 firms. Across all models, the F-statistics are highly significant at the 1% level, and the R-squared values range from 0.569 to 0.603, indicate acceptable model fit.
Regression results
Note: Standard errors in parentheses;
*** p < 0.001, ** p < 0.01, * p < 0.05.
Model 1 includes only the control variables and serves as the baseline for comparison. Several controls are significantly associated with risk-taking. TMT size (β = 0.003, p < 0.001) and TMT average tenure (β = 0.012, p < 0.001) are both positive and significant, while CEO change also has a positive effect (β = 0.017, p < 0.05). CEO experience is positive but not statistically significant (β = 0.013, ns). Among firm-level controls, slack resources (β = 0.283, p < 0.001) and operational capacity (β = 0.033, p < 0.001) are positively related to risk-taking. By contrast, asset turnover (β = –0.092, p < 0.001), and Tobin’s Q (β = –0.009, p < 0.01) are negatively related. R&D intensity also negative but not significant (β = –0.169, ns). Firm age also shows a positive association with risk-taking (β = 0.009, p < 0.001). Firm size was tested in robustness checks but remained non-significant and appears to be absorbed by other financial controls. We excluded it from reported models, as results were substantively unchanged.
Model 2 tests the direct effect of CEO origin on risk-taking. The coefficient for outsider CEOs is positive and significant (β = 0.035, p < 0.01), supporting Hypothesis 1. Relative to insider CEOs, outsiders are more likely to pursue risk-taking initiatives, consistent with the view that their limited embeddedness in firm-specific routines and norms affords greater discretion to depart from legacy trajectories and initiate bold, uncertain strategies. To ensure the results are not driven by the composite construction, we re-estimated the models using each component of risk-taking (R&D, CAPEX, and long-term debt) as separate dependent variables. In all cases, outsider CEOs remained positively and significantly associated with risk-taking, confirming that the findings are robust across different dimensions. These supplementary results are available upon request but are not reported for brevity.
Model 3 introduces short-termism as a moderator. The interaction between outsider CEO and short-termism is negative and significant (β = –0.028, p < 0.01), supporting Hypothesis 2a. This indicates that the positive effect of outsider CEOs on risk-taking diminishes under strong short-term temporal pressures, as such conditions constrain discretion and favor more cautious strategies.
Model 4 replaces short-termism with long-termism to test Hypothesis 2b. The interaction between outsider CEO and long-termism is positive and significant (β = 0.104, p < 0.001), consistent with the view that long-term horizons amplify outsider CEOs’ propensity to pursue risk-taking. Under long-termist conditions, outsider CEOs have greater latitude to implement bold initiatives and reposition firms toward future opportunities.
Model 5 incorporates attention breadth as a moderator. The interaction between CEO origin and attention breadth is positive and significant (β = 0.575, p < 0.05), supporting Hypothesis 3. This result suggests that outsider CEOs are better able to leverage broader attention to identify and act upon a wider range of strategic opportunities, thus enhancing firms’ risk-taking behaviors.
To further validate the hypothesized moderating effects, Fig. 2 to Fig. 4 present the interaction plots between CEO origin and the three behavioral moderators. Figure 2 shows the interaction plot between CEO origin and short-termism. Under low short-termism, outsider CEOs are associated with substantially higher levels of risk-taking compared to insiders. Under high short-termism, however, their positive effect is attenuated, with outsider CEOs pursuing less risk relative to the low short-termism condition. Insider CEOs’ risk-taking remains largely stable across temporal contexts. These results support Hypothesis 2a.
Interaction plot of CEO origin and short-termism.
Figure 3 presents the interaction plot for long-termism. Under high long-termism, outsider CEOs pursue significantly greater risk-taking compared to insiders, whereas under low long-termism, their risk-taking aligns more closely with insiders. This finding supports Hypothesis 2b.
Interaction plot of CEO origin and long-termism.
Figure 4 illustrates the interaction plot between CEO origin and attention breadth. Outsider CEOs with broad attention breadth exhibit elevated levels of risk-taking, while those with narrow attention breadth do not significantly increase risk-taking; insiders’ behavior remains relatively stable. This interaction supports Hypothesis 3. Taken together, the interaction plots visually reinforce the regression results, supporting our proposed models.
Interaction plot of CEO origin and attention breadth.
Discussions
Main findings
We examined how CEO origin influences firm-level risk-taking and how this relationship is conditioned by two behavioral moderators: temporal orientation (short-termism vs. long-termism) and attention breadth. Drawing on the microfoundations perspective, we theorized that outsider CEOs constitute a key individual-level input into strategic outcomes. Their effect on risk-taking is contingent on the behavioral context in which they operate, particularly the temporal orientations and cognitive framing that channel how strategic intent is enacted. The results provide empirical support for this integrative framework, clarifying the mechanisms through which CEO origin translates into bold, high-risk strategic behavior.
Our findings confirm Hypothesis 1: outsider CEOs are more likely to engage in strategic risk-taking than insider CEOs. This aligns with prior conceptualizations of outsider succession as a disruptive force, often associated with transformation mandates and strategic renewal (Firk et al., Reference Firk, Hennig, Meier and Wolff2024; Quigley et al., Reference Quigley, Hambrick, Misangyi and Rizzi2019). Because outsider CEOs are not embedded in existing routines and power structures, they bring less constrained cognitive frames and greater openness to uncertainty, which manifests in elevated levels of investment in R&D, capital expenditures, and long-term financing. Robustness checks confirm that this effect is not attributable to firm size or other financial characteristics but reflects outsiders’ distinctive orientations and discretion. These findings reinforce our theoretical framing of CEO origin not as a static demographic attribute but as a microfoundational mechanism through which firms reconfigure their cognitive infrastructure and strategic trajectory.
Several control variables exhibit significant effects. Asset turnover, which reflects operational efficiency, and Tobin’s Q, an indicator of market valuation, are consistently negative and significant. In contrast, R&D intensity remains non-significant. These findings suggest that firm risk-taking is not solely driven by higher R&D spending, efficiency, or valuation metrics. Instead, within-firm increases in efficiency may signal a shift toward exploitation-oriented strategies that constrain experimentation and innovation (Lavie, Stettner & Tushman, Reference Lavie, Stettner and Tushman2010). Moreover, the non-significance of R&D intensity could reflect the nature of asset-light firms, where growth and innovation are often pursued without extensive tangible investments (Huang, Ceccagnoli, Forman & Wu, Reference Huang, Ceccagnoli, Forman and Wu2013). Together, these patterns underscore that the observed effects of CEO origin and its interactions with cognitive moderators are not mere reflections of underlying financial conditions but rather stem from differences in executive cognition and strategic orientation.
Turning to the moderators, our findings confirm that the impact of outsider CEOs is context dependent. Supporting Hypothesis 2a, short-termism weakens the positive relationship between outsider CEO origin and risk-taking. In environments emphasizing near-term performance, outsider CEOs are less willing or less able to pursue uncertain long-term initiatives, as heightened monitoring and accountability limit their discretion. In contrast, consistent with Hypothesis 2b, long-termism strengthens the outsider CEO–risk-taking link. Under long-termist conditions, outsider CEOs enjoy greater latitude to enact bold strategies, pursue exploratory investments, and reposition firms for future opportunities. Together, these results underscore temporal orientation as a decisive contingency that channels whether outsider CEOs’ transformational mandate is constrained by short-term pressures or amplified by long-term horizons.
We also find strong support for Hypothesis 3: attention breadth strengthens the positive relationship between outsider CEOs and risk-taking. When outsider CEOs demonstrate broader attentional scope, they are significantly more likely to undertake risk-intensive actions. This aligns with arguments from the attention-based view that attentional capacity is a critical cognitive mechanism shaping how executives perceive, interpret, and respond to complex strategic environments (Durán & Aguado, Reference Durán and Aguado2022; Eklund & Mannor, Reference Eklund and Mannor2021). For outsider CEOs, broader attention facilitates opportunity recognition, reduces reliance on legacy heuristics, and expands the range of strategic options considered. Attention breadth thus compensates for informational disadvantages and empowers outsiders to act decisively under uncertainty.
Taken together, these findings demonstrate that CEO origin functions as a microfoundational antecedent of risk-taking, but its effect is contingent on behavioral moderators that shape how executive cognition is activated within organizational contexts. This supports the view that managerial judgment remains a critical mechanism through which firms navigate complexity and uncertainty (Foss & Klein, Reference Foss and Klein2023).
Our study makes three contributions. First, we advance the microfoundations literature by specifying how temporal orientation and attention breadth function as interface mechanisms that determine whether outsider CEOs’ risk-taking mandate is constrained or enabled (Cristofaro & Lovallo, Reference Cristofaro and Lovallo2022; Foss & Mazzelli, Reference Foss and Mazzelli2025). Second, we help reconcile mixed findings on outsider succession and firm performance (Bromiley & Rau, Reference Bromiley and Rau2016; Hambrick & Lee, Reference Hambrick and Lee2025), showing that outcomes depend not only on CEOs’ structural detachment but also on the behavioral conditions shaping their discretion. Third, we contribute to a more nuanced understanding of CEO influence by clarifying when and how outsider leaders can realize their strategic potential. These insights underscore the importance of behavioral alignment following CEO succession and call for greater attention to the contextual forces that modulate executives’ impact on strategic outcomes.
Theoretical implications
We offer several theoretical contributions to the literature on firm risk-taking by applying a microfoundations perspective to CEO behavior and decision contexts. First, we extend understanding of firm risk-taking by identifying CEO origin particularly outsider status as a central antecedent. Prior research has often treated CEO origin as a structural classification within succession studies (Chen & Hambrick, Reference Chen and Hambrick2012). Our findings build on this literature by showing that the strategic consequences of outsider appointments are rooted in cognitive and behavioral differences, not merely in formal role transitions. Outsider CEOs differ in how they process information, interpret the firm’s environment, and approach high-stakes decisions. In this way, CEO origin operates as a necessary lever (Hambrick & Lee, Reference Hambrick and Lee2025; Quigley et al., Reference Quigley, Hambrick, Misangyi and Rizzi2019): boards can use succession choices to deliberately reconfigure a firm’s cognitive infrastructure and strategic trajectory. From a microfoundations perspective, we assert that CEO origin is more than a demographic attribute, it functions as an individual-level input that shapes strategic behavior through cognition, attention, and intent (Cristofaro & Lovallo, Reference Cristofaro and Lovallo2022; Felin & Foss, Reference Felin and Foss2005; Heubeck & Meckl, Reference Heubeck and Meckl2023). This insight expands the conventional research on CEO origin.
Second, we show that temporal orientation conditions the effect of CEO origin on firm risk-taking. Short-termism constrains the translation of executive intent into bold action by narrowing decision horizons and heightening sensitivity to immediate performance pressures, thereby suppressing risk-taking (Agnihotri et al., Reference Agnihotri, Bhattacharya and Satya Prasad2025; Marginson & McAulay, Reference Marginson and McAulay2008). By contrast, long-termism broadens the evaluative horizon, legitimizing delayed payoffs and enabling outsider CEOs to enact transformative strategies. Moreover, we also show that attention breadth conditions outsider CEOs’ propensity for risk-taking. Broader attentional scope facilitates exploration by widening cognitive search, improving opportunity recognition, and enabling the integration of diverse informational cues (Brielmaier & Friesl, Reference Brielmaier and Friesl2023; Eklund & Mannor, Reference Eklund and Mannor2021). Together, these insights delineate the boundary conditions under which outsider CEOs pursue risk-taking, helping explain the variability observed in prior studies of succession (Bromiley & Rau, Reference Bromiley and Rau2016; Quigley et al., Reference Quigley, Hambrick, Misangyi and Rizzi2019) and strategic risk-taking (Arrfelt et al., Reference Arrfelt, Mannor, Nahrgang and Christensen2018).
Third, we contribute to the microfoundations literature by demonstrating how individual-level characteristics, such as CEO origin, interact with organizational conditions to shape firm-level outcomes. In line with recent calls to specify cross-level mechanisms (Arndt et al., Reference Arndt, Galvin, Jansen, Lucas and Su2022; Felin et al., Reference Felin, Foss and Ployhart2015; Foss & Mazzelli, Reference Foss and Mazzelli2025), our results show that temporal orientation and attention breadth act as behavioral filters that condition the expression of executive cognition. Short-termism restricts the translation of outsider CEOs’ transformational intent into bold strategies, whereas long-termism and attention breadth provide enabling contexts that amplify risk-taking. By modeling these situational pathways, we move beyond simple trait–outcome associations and capture how executive-level variation is enacted within bounded organizational contexts. This provides a concrete operationalization of how microfoundations unfold dynamically, reinforcing efforts to bridge individual dispositions, contextual moderators, and organizational behavior (Cristofaro et al., Reference Cristofaro, Augier, Lovallo, Abatecola and Leoni2024; Cristofaro & Lovallo, Reference Cristofaro and Lovallo2022; Palmié et al., Reference Palmié, Rüegger and Parida2023).
Finally, we extend the view of CEO succession as a strategic investment under uncertainty (Hambrick & Lee, Reference Hambrick and Lee2025; Quigley et al., Reference Quigley, Hambrick, Misangyi and Rizzi2019). Our findings suggest that the value of outsider appointments is not inherent, but contingent on the internal conditions that support or suppress executive discretion. Outsider CEOs are more likely to catalyze firm risk-taking in environments characterized by long-term horizons and broad cognitive scope, and less likely to do so in environments dominated by short-term pressures and narrow attentional focus. Succession should therefore be understood as a strategic act that reconfigures the cognitive infrastructure of the firm, with outcomes dependent on post-succession alignment between executive intent and organizational context.
Managerial implications
Beyond theoretical contributions, our findings offer practical insights for boards, executive recruiters, and senior leaders involved in CEO succession and strategic transformation. First, appointing an outsider CEO should not be viewed as merely symbolic. Such appointments can catalyze risk-taking when the internal environment provides supportive conditions. Outsider CEOs often bring fresh perspectives and a stronger appetite for risk, but their effectiveness ultimately depends on the behavioral context they face. Second, firms pursuing innovation, R&D growth, or strategic renewal should carefully evaluate their temporal orientation. In short-termist environments, even bold outsider CEOs may be forced to prioritize immediate results, limiting their ability to pursue high-variance strategies. By contrast, long-termist contexts expand discretion, legitimize delayed payoffs, and empower outsider CEOs to undertake riskier initiatives that reposition the firm for future opportunities. Boards can reinforce these conditions by aligning incentives, performance evaluations, and governance systems with long-term goals. Third, attention breadth emerges as a critical enabler of outsider CEO effectiveness. Organizations can foster this by designing onboarding and leadership development programs that expose new leaders to diverse strategic domains, such as cross-functional briefings, temporary rotations, or early participation in multi-unit planning processes. These practices help CEOs broaden their cognitive frames, accelerate sensemaking, and navigate unfamiliar organizational environments more effectively. Finally, succession should be understood not only as a leader-centric decision but also as a matter of leader–context alignment. Boards should evaluate both CEO qualifications and the organization’s temporal and cognitive readiness for change. By assessing internal time horizons and attentional structures, boards can improve fit and increase the likelihood that succession decisions translate into successful strategic transitions.
Limitations and directions of future research
Despite its contributions, this study has several limitations that offer avenues for future research. First, our use of textual analysis provides a structured method for capturing executive cognition, it relies on indirect proxies derived from shareholder letters. These documents may not fully reflect internal deliberations or informal cognitive processes. Future research could strengthen construct validity by triangulating text-based measures with interviews, surveys, or internal communications. Second, our sample of large U.S.-based firms enhances comparability but restricts generalizability. Investigating different contexts such as smaller firms, family-owned businesses, or organizations in emerging markets, may reveal additional contingencies and behavioral patterns among CEOs. Third, our binary classification of CEO origin (insider vs. outsider) does not capture the full range of executive career pathways. Future studies could refine this measure by distinguishing among external hires with varying industry backgrounds, prior tenure lengths, or reappointment histories. Finally, we use fixed-effects models to mitigate unobserved heterogeneity, endogeneity concerns cannot be fully ruled out. Factors such as board dynamics, succession planning, or prior performance shocks could simultaneously influence both CEO selection and risk-taking behavior. Future research could leverage instrumental variables to better isolate causal effects.
Conclusion
This study contributes to the understanding of strategic risk-taking by examining how outsider CEO origin interacts with internal behavioral conditions, specifically temporal orientation and cognitive focus. Drawing on a 20-year panel of 100 S&P 500 firms, we find that outsider CEOs are generally more inclined toward risk-taking than insiders, but this effect is contingent. Short-termism dampens their tendency to take risks, while long-termism and broad attention amplify the tendency. By incorporating these internal moderators, we move beyond static categorizations of CEO origin and emphasize the temporal and cognitive environments that condition executive influence. This aligns with recent calls in the microfoundations literature to specify how individual traits interact with organizational context (Durán & Aguado, Reference Durán and Aguado2022; Felin et al., Reference Felin, Foss and Ployhart2015; Foss & Mazzelli, Reference Foss and Mazzelli2025). In sum, outsider CEO origin is not an inherent driver of strategic boldness; its effects depend on the behavioral landscape in which executives operate. By highlighting the moderating roles of temporal orientation and cognitive focus, our study provides a more nuanced account of CEO behavior and underscores the importance of aligning leadership choices with organizational context. We invite future research to further investigate the contextual mechanisms that shape executive decision-making, ensuring a deeper understanding of when and how CEOs can act as catalysts for strategic change.
Appendix A1. Dictionaries for Temporal Orientation
Source: Brochet et al. (Reference Brochet, Loumioti and Serafeim2015).
Appendix A2. Dictionaries for 13 Strategic Issue Categories
Note: * is used to find variations of the searching word.
Source: Eklund and Mannor (Reference Eklund and Mannor2021).
Open access
