Getting by with the advice of their friends: CEOs' advice networks and firms' strategic responses to poor performance.

A number of interconnected literatures have sought to explain the robust finding from research in organization studies and strategic management that top executives often do not initiate significant changes in corporate strategy in response to poor firm performance and, paradoxically, may adhere even more firmly to current strategies in response to performance problems. A central theme in much of this literature is that systematic biases or "perceptual distortions" among top managers are at least partly responsible for strategic inertia in the face of poor performance (Starbuck, Greve, and Hedberg, 1978:113; Whetten, 1980, 1987; Hambrick and D'Aveni, 1988; for a review, see Barker and Duhaime, 1997). Top managers are thought to overattribute poor performance to external factors, such as a competitive industry environment, and underattribute performance problems to weaknesses in their current strategy, even when competitors have performed better under the same industry conditions.

Threat-rigidity theory suggests that stress brought on by poor performance prompts executives to reduce information processing activity, which decreases their consideration of strategic alternatives (Staw, Sandelands, and Dutton, 1981; Sutton, 1990). This restriction in information processing is attributed in part to the centralization of decision making in response to poor performance and an associated decrease in communication between lower-level managers and top executives. The organizational decline literature also associates strategic inertia with centralized decision-making processes (Whetten, 1987). But empirical research has provided mixed support for these propositions. Large-sample studies of letters to shareholders and other corporate communications, as well as qualitative studies of organizations in decline, have provided evidence for managers' attribution biases and restricted consideration of strategic alternatives during periods of poor performance (Starbuck, Greve, and Hedberg, 1978; Salancik and Meindl, 1984; D'Aveni and MacMillan, 1990; Barker and Bart, 2002). There is less evidence that centralized decision making contributes to strategic inertia in response to poor performance (Cameron, Whetten, and Kim, 1987; D'Aveni, 1989; Barker and Mone, 1998). While this literature has significantly advanced our understanding of strategic inertia by demonstrating how managers' cognitions can help explain firms' responses (or lack thereof) to economic adversity, important gaps in our understanding of this phenomenon remain. Especially noteworthy is that there has been little consideration of how micro-social factors may contribute to strategic inertia. In particular, despite the recent explosion of interest in how social networks can influence corporate policy (e.g., Haunschild and Miner, 1997; Davis and Greve, 1997; Ingram and Roberts, 2000), there has been little research on the role that executives' social networks might play in shaping firms' strategic responses to poor performance (cf. Kraatz, 1998) as they seek advice through their social interactions with other top managers that will give them qualitative assessments of current firm strategy.

Research on strategic decision making and executive scanning has shown that executives assign greater weight to information and advice from personal sources, such as informal conversations with colleagues, than to impersonal sources, such as written reports and recommendations or the output of management information systems, in making strategic decisions (Aguilar, 1967; Mintzberg, 1973; Daft, Sormunen, and Parks, 1988; Brown and Eisenhardt, 1997; Elenkov, 1997). This literature would suggest that advice-seeking interactions with colleagues may have a significant influence on chief executive officers' decisions about whether to modify strategy in response to performance problems.

Moreover, social psychological research on belief perseverance suggests how CEOs' advice seeking in response to poor firm performance may contribute to the previously noted biases in managers' perceptions that have been implicated in strategic inertia and organizational decline. A large body of research has shown that people often persevere in their beliefs even when the evidential basis for those beliefs has been largely disconfirmed, consistent with research on organizational decline and threat rigidity, suggesting that executives' confidence in their strategies often persists despite negative performance outcomes (Nisbett and Ross, 1980; Anderson and Kellam, 1992). Such belief perseverance has been shown to result not only from biased assimilation of available evidence but also from biased search for information and opinion (Nisbett and Ross, 1980; Carretta and Moreland, 1982; Hastie, Penrod, and Pennington, 1983; Schulz-Hardt et al., 2000). Research in this literature has shown that when people's beliefs are challenged, they tend to seek information from sources that are likely to affirm those beliefs, particularly personal sources of information, and avoid sources that are more likely to disconfirm those beliefs (Swann, 1996). Recent theoretical interpretations of these findings have invoked self-categorization theory (Hogg and Terry, 2000), suggesting that uncertainty created by evidence that calls one's preexisting beliefs into question may result in increased information seeking from in-group members who are likely to provide points of view that affirm those beliefs.

The uncertainty that results from relatively poor firm performance may prompt CEOs to seek more advice from executives at other firms whom they would categorize as in-group members because they share a common professional background, friendship ties, or employment in the same industry. Evidence from related social psychological and social network research indicates that these in-group managers are especially likely to offer information and points of view that affirm CEOs' preexisting strategic beliefs, defined as their beliefs about what firm strategies are likely to succeed given prevailing environmental demands. CEOs' strategic beliefs are likely to be instantiated to a significant degree in their firms' current strategies (Finkelstein and Hambrick, 1996), and when executives' advice seeking restores their confidence in the correctness of their strategic beliefs, they will be less likely to change firm strategy. In the study presented here, we use survey data on top executives' advice-seeking interactions to help explain strategic inertia in response to relatively poor firm performance, thus providing insight into the role that executives' social networks could play in firms' strategic responses to economic adversity.

CEOS' ADVICE NETWORKS AND FIRMS' STRATEGIC RESPONSES TO POOR PERFORMANCE

Recent social psychological theorizing about human motivation points to the desire to render the environment understandable and predictable as among the most basic drivers of individual cognition and behavior (Stevens and Fiske, 1995; Fiske, 2000), primarily because understanding serves the even more basic need to exercise control over personally important outcomes (Pittman, 1998). We use the term "subjective uncertainty" to refer to the psychological state that exists when an individual is "confronted with an experience that calls his or her conceptualizations into question, that implies that his or her understanding of the world is inadequate ..." (Pittman and D'Agostino, 1985: 120; cf. Hogg and Abrams, 1993; Hogg and Mullin, 1999). Because subjective uncertainty suggests a reduced capacity to control important outcomes, it evokes psychological distress and negative affect (e.g., anxiety), and this activates a deep-seated motive to restore a homeostatic level of certainty and perceived control (Pittman and Heller, 1987). Evidence suggests that people pursue a wide range of cognitive and behavioral strategies in an effort to reestablish a sense of certainty and control (Greenberger and Strasser, 1991). To the extent that CEOs' judgments and beliefs about strategy are reflected to some degree in their firm's current corporate strategy (see Finkelstein and Hambrick, 1996), poor performance will reduce a CEO's confidence in the veracity of his or her beliefs about strategic cause-and-effect relationships and prompt efforts to reduce uncertainty. In suggesting that poor performance raises CEOs' subjective uncertainty about their strategic beliefs, however, we are assuming neither that poor performance discredits CEOs' beliefs and assumptions about strategy with absolute certainty, given inherent ambiguity in determining the effect of corporate strategy on firm performance (Pfeffer, 1981), nor that the determinants of poor performance are so ambiguous that CEOs can completely discount the negative feedback. Under such circumstances, CEOs would become more uncertain about their strategic beliefs following poor performance and yet could be reassured about those beliefs through interacting with other managers.