essay

r Academy of Management Journal 2015, Vol. 58, No. 6, 1740–1760. http://dx.doi.org/10.5465/amj.2012.1091

PAYING THE PRICE? THE IMPACT OF CONTROVERSIAL GOVERNANCE PRACTICES ON MANAGERIAL REPUTATION

MICHAEL K. BEDNAR E. GEOFFREY LOVE MATTHEW KRAATZ University of Illinois

This study directly examines the reputational penalties that managers pay when they engage in controversial governance practices that raise questions about managerial self- interest. These penalties should deter questionable behavior and enable reputation to serve a social control function, yet we know little about how and when these penalties are actually imposed. Unlike prior research in this vein, we account for the fact that reputational penalties associated with such practices may differ across audiences be- cause of differences in interpretations of the practice and differences in causal attri- butions about its use. Specifically, we develop theory to explain how and when stock analysts and peer executives applied reputational penalties to managers when firms used a poison pill, a prominent anti-takeover device. We find that the reputational penalties associated with poison pills differed substantially between these two groups and that these groups applied different penalties depending on the media coverage that the poison pill received, the performance of the firm, and the extent to which the practice had already been adopted. The findings suggest that reputational penalties for questionable behaviors may be more contingent and harder to sustain than previously thought.

Governance decisions often attract controversy when important groups perceive them as being mo- tivated by managerial self-interest rather than the good of the firm. But what is the reputational price that managers of a firm pay for such decisions? This question is an important one for both reputation and governance research, because it cuts to the core of reputation’s ability to deter questionable behaviors and thus serve as a social control mechanism. A re- curring theme within the reputation literature is that the threat of reputational penalty (i.e., a loss of rep- utation) can discourage uncooperative behaviors and promote exchange (Burt, 2005; Coleman, 1988; Klein & Leffler, 1981). This same basic logic is seen in the corporate governance literature. An underlying assumption in governance research from an agency perspective is that managers are motivated to avoid

damage to their valuable reputations by eschewing behavior that will be viewed as self-serving or oth- erwise opposed to owner interests (Dalton, Hitt, Certo, & Dalton, 2007; Fama & Jensen, 1983).

However, despite these potential penalties, firms regularly implement governance practices that raise concerns about managerial self-interest (Greve, Palmer, & Pozner, 2010). Some ready examples in- clude anti-takeover devices (Davis, 1991), CEO duality (Finkelstein & D’Aveni, 1994), and compensation pol- icies that reward managers even when they underper- form (Bebchuk & Fried, 2006). The prevalence and persistence of such practices raises questions about the reputational penalties supposedly associated with them. As such penalties are central to reputation-as- social-control arguments, questions about when these penalties are applied, who applies them, and why they do so, are important to consider if we are to make sense of the role reputation plays where controversial gov- ernance practices are concerned. However, to date, empirical evidence regarding these penalties is lack- ing. Moreover, as we discuss further below, extant theory regarding reputation’s role in governance ap- pears somewhat oversimplified, in part because it fails to account for differences between evaluating audiences, as well as the complexity of the attribution process that leads to reputational penalties.

We thank Tim Pollock and three anonymous reviewers for their constructive feedback and guidance throughout the review process. We are indebted to Jerry Davis for his generous sharing of data. This manuscript benefitted from the helpful comments of Ruth Aguilera, Steve Boivie, Jeff Loewenstein, Joe Mahoney, and seminar participants at the University of Michigan, the University of Virginia, the Midwest Strategy Meeting, and the BYU/University of Utah Winter Strategy Conference.

1740

Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyright holder’s express written permission. Users may print, download, or email articles for individual use only.

Our study takes a step to address these issues by developing and testing theory about the reputational penalties associated with controversial governance practices. We start by recognizing that managers are sensitive to the views of different stakeholders, as well as those of managers at peer firms (Davis, 1991; Fombrun, 1996; Mitchell, Agle, & Wood, 1997). These groups represent distinct audiences that may interpret the same questionable practice in different ways and apply different reputational penalties. Yet, work on reputation’s social control capabilities has typically emphasized contexts where reputational evaluators constitute a single, relatively homoge- neous audience (e.g., close-knit merchant communi- ties) (Coleman, 1988; Granovetter, 1985). Similarly, much corporate governance literature appears rather one-sided, as it tends to emphasize the reputational judgments of owners rather than also considering those of other important groups (e.g., employees, customers, other managers) (Fama & Jensen, 1983). A clear implication is that we need to develop theory that accounts for different audiences and explains how the same practice may affect the reputations of the firm’s managers differently, depending on which audience makes the evaluation.

To do this, we conceptualize reputation as a social judgment, wherein evaluators interpret other actors’ behaviors and use these as indicators of the actors’ underlying dispositional traits (i.e., character, com- petence) (Bromley, 1993; Fiske & Taylor, 2013). In the context of controversial governance practices, this suggests that different groups may interpret a practice based on the particular governance models to which they subscribe, and assign character traitsto the managers involved. One aspect that appears to be underemphasized in this work, however, is the way that causal attributions (Kelley, 1967) intervene in this process. Even if two groups interpret a contro- versial practice in similar ways, that practice’s rep- utational impact could still vary dramatically if one group attributes its use to managers’ self-serving character (i.e., an internal attribution leading to a more severe penalty) whereas another sees it as a forced response to a threat (i.e., an external attribu- tion leading to a weaker penalty). Such a situation could arise, for example, if the evaluating groups differ in whether their attributions are based on in- group or out-group perspectives (Hewstone, Rubin, & Willis, 2002; Pettigrew, 1979). The possibility of such differences in attributions, and thus differences in reputational penalties across groups, is a second missing piece in extant discussions of links between questionable behaviors and reputational penalties.

We further develop these ideas about the influence of different audiences and different attributions on reputational penalties, and test them through an empirical study of how the reputation of a firm’s managers are influenced by a particular contro- versial governance practice: poison pills, a prom- inent anti-takeover device. Because this practice has diffused widely across corporate America, it provides an attractive opportunity to examine what these rep- utational penaltiesactuallyare,whentheyareapplied, and by which groups. We focus on two evaluating audiences that have been widely discussed in the governance literature: stock analysts, who are often portrayed as owner surrogates (Zuckerman, 1999); and fellow managers at peer firms, who clearly influence othermanagersingovernancedecisions(McDonald & Westphal, 2003). We hypothesize that these two groups will interpret poison pills in systematically different ways and make distinctive causal attribu- tions about their use, resulting in differences in the reputational penalties that they apply. Specifically, we draw on attribution theory (Heider, 1958; Jones & Davis, 1965; Kelley, 1967), as well as intergroup bias research (Hewstone et al., 2002), to argue that analysts will tendto make internal attributionsaboutthe use of poison pills and apply more severe penalties. On the other hand, peer managers will tend to attribute poi- son pills’ use externally, to situational pressures, and so apply weaker penalties. We also consider several contextual factors that have been discussed in both the reputation and governance literatures, including media coverage of the practice, firm performance, and the extent of prior adoptions. We propose that these factors will affect the attributions that analysts and peer managers make in different ways, further differentiating the reputational penalties imposed by these groups. These contingencies also help us to shed light on when the reputational penalties asso- ciated with governance practices are likely to be stronger or weaker in general.

This study contributes to both the corporate gover- nance and reputation literatures in two ways. First, by directly assessing the reputational penalties that managerspaywhentheirfirmsengageincontroversial governance practices, our study tests an important assumption within the governance literature—that managers willsuffer reputational damage for engaging inpracticesthatcanbeseenasself-serving.Second,by accounting for differences between evaluating audi- ences, we move away from a one-sided view of repu- tational penalties in governance by providing both a theoretical and empirical basis for considering how different audiences interpret the same controversial

2015 1741Bednar, Love, and Kraatz

practice. Third, we build a more socialized perspec- tive regarding how reputation operates in the gover- nance context, which accounts for complexities of interpretation and subsequent attributions that are likely to shape reputational penalties, but that are missing from existing generalized treatments of repu- tation found in signalingand agency theory. Our study also has implications for the broader reputation liter- ature, as governance is likely not the only context wherein such complexities influence the reputational penaltiesassociatedwithquestionablebehaviors.And while audience-specific reputations and attribution processes are established ideas within the reputation literature(e.g.,Kim&Jensen,2014),theyhavenotbeen considered together and their implications have been less developed with respect to reputation’s social control function.

REPUTATION, GOVERNANCE, AND SOCIAL CONTROL

We conceptualize managerial reputation as a per- ceptual judgment of the manager’s overall role performance, as assessed by a specific audience (Graffin, Pfarrer, & Hill, 2012; Tsui, 1984). In general, reputations are seen to arise from evaluations of an actor’s behaviors and outcomes over time (Bromley, 1993; Weigelt & Camerer, 1988). As part of this pro- cess, people tend to attribute underlying traits to others that reflect their apparent competence and character (Fiske & Taylor, 2013; Fiske, Cuddy, & Glick, 2007). An actor who regularly meets com- mitments or demonstrates exceptional capabilities will gain a positive reputation, but if the actor be- haves opportunistically or performs below expecta- tions, he or she is likely to pay a reputational price (Granovetter, 1985; Klein & Leffler, 1981). For man- agers in particular, strong reputations have been linked with significant benefits, including higher performance ratings, more rapid promotion, and higher compensation (Burt, 2007; Graffin et al., 2012; Tsui, 1984). Such benefits arise because an actor’s future behaviors are often not knowable in advance of an interaction, so people reduce uncertainty by looking to the actors’ reputation as a predictor thereof (Bromley, 1993; Weigelt & Camerer, 1988). Positive reputations lead others to see the actor as an attractive exchange partner, while negative ones impede exchange (Klein & Leffler, 1981). It is well established that these features can enable reputation to provide a mechanism for social control, in that it deters selfish actions and aligns the interests of ex- change partners (Coleman, 1988; Shapiro, 1983), and

thus enables exchange and productive interaction (Burt, 2005). A number of researchers have argued that actors are motivated to treat trade partners fairly because if a dishonest act is detected, the offender’s good reputation will be lost as word quickly spreads through the community and future trade opportu- nities are diminished (Coleman, 1988; Granovetter, 1985; Uzzi, 1999).

The basic logic underlying reputation’s potential social control function also appears in the corporate governance literature (Ertimur, Ferri, & Maber, 2012; Fama & Jensen, 1983; Gomes, 2000). A central focus therein has been on the agency problem, which ari- ses from potential conflicts of interest between managers (agents) and owners (principals), and that is manifested when managers take self-serving ac- tions rather than furthering the best interests of owners (Dalton et al., 2007; Jensen & Meckling, 1976). Reputation enters in because managers are assumed to be motivated to preserve and enhance their reputations by avoiding behavior that could be viewed as self-serving or opportunistic (Fama & Jensen, 1983). Managers whose actions appear to reveal negative character traits can expect damage to their reputations, which we conceptualize as a rep- utational penalty that is imposed by evaluating audiences. These reputational penalties can have significant consequences for managers with respect to a variety of tangible outcomes; for example, through the “settling up” mechanism, which leads to reduced opportunities in the labor market or reductions in compensation (Arthaud-Day, Certo, Dalton, & Dalton, 2006; Cowen & Marcel, 2011; Wowak & Hambrick, 2010).

REPUTATIONAL PENALTIES AND CONTROVERSIAL GOVERNANCE PRACTICES

Corporate governance provides a rich domain to examine reputational penalties (i.e., damage) be- cause of the number of governance-related practices that attract controversy and have the potential to send negative signals about management. Some of these practices are related to compensation agree- ments. For example, managerial perquisites such as the use of corporate jets (Boivie, Lange, McDonald, & Westphal, 2011), high compensation (Core, Guay, & Larcker, 2008; Wade, O’Reilly, & Pollock, 2006), golden parachutes (Wade, O’Reilly, & Chandratat, 1990), and other arrangements that result in lucra- tive outcomes for managers regardless of firm performance often lead to negative character attri- butions of the managers involved. Another category

1742 DecemberAcademy of Management Journal

of controversial governance practices is related to managerial power. For example, when CEOs also serve as chairmen of the board (Finkelstein & D’Aveni, 1994) or when the board has many in- siders, such practices are often attributed to man- agers seeking the ability to act without appropriate oversight and scrutiny. A final category has to do with the market for corporate control and managerial entrenchment. Many anti-takeover provisions, such as poison pills, are often seen as attempts to protect managerial job security rather than furthering the interests of the firm (Davis, 1991; Gompers, Ishii, & Metrick, 2003).

While these practices vary in important ways, they share a central feature in that they raise concerns and controversy because the managers involved can ap- pear to be acting in a self-serving manner. It is im- portant to note that none of these practices are illegal, so even if they raise questions about the character of the managers involved in their implementation, they can be, and are, justified in various ways by those same managers (Hirsch, 1986). That said, if reputa- tional evaluators apply the basic logic described above, these practices would seem to be just the type that we would expect to produce significant reputa- tional penaltiesfor managers thatchoosetousethem.1

Aswementioned above, however, these practicesand the broader governance context pose complexities that have not been theoretically developed in existing work. In particular, we suggest that existing models of reputation as social control require two important adaptations if we are to understand the reputational penalties associated with controversial governance practices. First, we must account for different audi- ences, since different groups may interpret the same practiceindistinctways.Second,wemustaccountfor variations in the causal attributions made by different audiences and in different circumstances.

Different Audiences

It is well established that managers respond to the needs and demands of different types of stake- holders, as well as the decisions and opinions of

managers at peer firms (Davis, 1991; Mitchell et al., 1997). This would seem to imply that managers are also sensitive to these groups’ reputational assess- ments of them. Yet, although the broader reputa- tional literature has discussed reputation as an audience-specific social evaluation that may differ across groups (Kim & Jensen, 2014), this has not been emphasized in discussions of reputation as social control. Instead, such work has tended to conceptu- alize reputational evaluators as relatively homoge- neous groups (e.g., workgroups, friends, close-knit communities) (Coleman, 1988; Granovetter, 1985). Similarly, in the governance literature, much re- search has been based on an agency view, which tends to take the perspective of the shareholder when assessing the appropriateness of governance prac- tices (Fama & Jensen, 1983). However, taking account of differences in the reputational consequences of managerialdecisionsseemsparticularlyimportantin a governance context, because audiences may dif- fer in the models of governance that shape their evaluations of governance practices. Although the agency view looks to the interests of shareholders in determining what constitutes good governance (Dalton et al., 2007; Zajac & Westphal, 2004), a prominent alternative view holds that good gover- nance is primarily concerned with acting in the best interests of the firm as an ongoing entity. This view, which has a long history and which remains prom- inent in the managerial community, among others (Blair & Stout, 1999; Freeman & Evan, 1990), implies that managers need to take a more expansive view of which stakeholders need to be considered when im- portant decisions are made. This latter perspective is often visible in legitimating accounts of controversial governance practices (Davis & Greve, 1997; Hirsch, 1986),andsuggeststhatimportantaudiencesmaynot view these as an unambiguously negative signal. These points highlight the need to account for these multiple audiences if we are to understand the rep- utational penalties associated with controversial governance practices.

Different Attributions

Our second adaptation of existing theory relates to the idea that the reputational consequences of con- troversial governance practices are likely to revolve around the causal attributions that evaluators make regarding managers’ motivations for their use. As mentioned above, evaluators make reputational as- sessments in part by interpreting behaviors as in- dicators of an actor’s underlying character traits

1 While some research and practices focus on individual managers (e.g., the CEO), discussions of governance prac- tices often refer to top managers at the firm, rather than a single manager. Consequently, our use of the term “managerial reputation” should be taken to refer to eval- uations of the collective management of the firm. We also distinguish managers’ reputations from those of the firms that they lead (Graffin et al., 2012).

2015 1743Bednar, Love, and Kraatz

(Bromley, 1993; Fiske & Taylor, 2013). When evalu- ators perceive apparent character flaws, this elicits particularly strong reactions (Fiske & Taylor, 2013). In the case of controversial governance practices, the possibility of reputational damage arises in large part because these practices can be seen as indicators of a manager’s self-serving nature. However, whether or notanevaluatormakessuchanegativetraitattribution depends largely on whether the evaluator makes an internal or external causal attribution (Kelley, 1967) regarding the manager’s motivations for use of the practice. For instance, when evaluators make in- ternal attributions about the presence of a contro- versial practice (e.g., that it is due to managerial self-interest), managers are likely to suffer a more se- vere reputational penalty, but if an audience makes external attributions (e.g., that situational factors drove its use), the penalty is likely to be muchsmaller.

While these points suggest the importance of ac- counting for causal attributions, they have not been emphasized in models of reputation in governance. They have, however, long been of central impor- tance in attribution theory itself, which deals with how people make sense of the world around them by making inferences about the cause of observed behaviors (Heider, 1958; Kelley, 1967; Kelley & Michela, 1980). This work provides some clues re- garding how causal attributions may differ across audiences. When an action has low social desir- ability or violates expectations, for instance, eval- uating audiences are more likely to attribute the action to internal dispositional traits of the actor (Jones & Davis, 1965). This implies that if different groups have different expectations of actors, or ad- here to different models of appropriate behavior, the groups will vary in their likelihood of attributing actions to internal dispositional traits.

We have also alluded to the idea that attributions can depend not just on attitudes toward a specific behavior, but also on the evaluators’ social position vis-à-vis the actor. A large body of work examining intergroup bias has shown that people systemati- cally tend to assess those in their own membership group (the in-group) more favorably than those who fall outside of this group (out-groups) (Hewstone et al., 2002; Mackie & Smith, 1998). Relatedly, eval- uators are more likely to attribute an out-group member’s negative actions to internal dispositional traits, while in-group members are more likely to look to situational factors to explain negative be- havior (Pettigrew, 1979; Schruijer et al., 1994). Ac- counting for in-group and out-group influences seems particularly important where controversial

governance practices are concerned, as group in- terests and identities are often salient in these cases. Outside parties who tend to use an out-group per- spective may impose different reputational penalties on the managers involved compared to audiences whose social position leads them to apply an in- group lens (e.g., those internal to the firm, or man- agers at similar firms).

CONTEXT AND HYPOTHESES: AUDIENCES, POISON PILLS, AND REPUTATIONAL

PENALTIES

We now apply these general ideas regarding audi- ences and attributions to our more specific research context. In our empirical study, we focus on two key evaluating audiences: stock analysts and peer exec- utives. Analysts act as information intermediaries in financial markets and their views are generally thought to represent those held by the investment community (Zuckerman, 1999). Managers are known to view the investment community, including ana- lysts, as a highly legitimate and powerful stakeholder group that can affect their job security and other im- portant outcomes (Davis, 2009; Mitchell et al., 1997; Useem, 1993; Wiersema & Zhang, 2011). Peer-firm executives also represent an important group, as seen in research showing that managers are quite sensitive to these actors’ evaluations of actions (Davis, 1991), and in work arguing that top managers constitute a cohesive social group with a sense of shared in- terests and outcomes (Useem, 1984). In addition, while peer executives are not typically viewed as a stakeholder group, their opinions can have tangible consequences for managers at the focal firm, such as affecting an executive’s standing in the broader managerial labor market.

In general, these two groups are seen as sharing a common perspective on many strategic and finan- cial issues, and are known to assess firms’ reputa- tions in quite similar ways (Fombrun & Shanley, 1990). However, it is well known that these groups have taken different positions regarding many of the controversial governance practices mentioned above. Since the 1980s, investors’ interpretations of governance issues have tended to be driven by the agency-logic and shareholder-value models of cor- porate governance (Zajac & Westphal, 2004). At the same time, investors have long been constituted as a group that is external to the firm’s management and operations (Berle & Means, 1932). This suggests that members of the investment community will tend to evaluate managers using an out-group perspective,

1744 DecemberAcademy of Management Journal

which will push them toward more negative assess- ments and internal causal attributions when concerns are raised (Pettigrew, 1979). Analysts in particular are likely to evaluate managers from an out-group perspective, as they are charged with scrutinizing management’s actions and can thus be seen as playing a “watchdog” role for investors in general (Coffee, 2006).

On the other hand, peer executives are more likely to interpret governance practices in a more nuanced fashion. While theytend to respect the long- established view that management decisions should advance the best interests of the firm and its multiple stakeholders (Blair & Stout, 1999), they are still likely to be influenced by the shareholder-value model’s ascendance (Zajac & Westphal, 2004). Managers can also be quite sensitive to the appearance of entrenchment and self-interest that controversial practices can present. Additionally, executives can feel a sense of group identity with their peers (Useem, 1984) and often experience similar pres- sures to them, suggesting that they are likely to apply an in-group logic when evaluating fellow managers. This should lead to more positive evaluations and a tendency to attribute other managers’ use of con- troversial practices to external contingencies and pressures.

As mentioned above, the practice we specifically focus on is that of the poison pill, a controversial takeover defense that arose in the mid-1980s in re- sponse to a wave of hostile takeovers (Davis, 1991; Davis & Greve, 1997). A poison pill is intended to make a firm an unattractive takeover target by granting shareholders of a target firm the right to buy shares of the newly formed entity at a deeply dis- counted price in the event of a hostile takeover not approved by the board. Poison pills were typically regarded as the most effective anti-takeover device, and were at the center of a controversy around takeovers and defenses against them (Jensen, 1988). This controversy centered on two distinct views of anti-takeover devices. From one perspective, anti- takeover maneuvers were seen as an appropriate response to the threat that hostile takeovers pre- sented to the firm as an ongoing entity, and to stakeholders such as employees and communities (Hirsch, 1986). Alternatively, these practices could be interpreted as unjustified interference with the market for corporate control’s function of enabling investors to replace under-performing managers (Jensen, 1989). Yet, in spite of the controversy around them, poison pills spread widely and rap- idly. The first poison pill was adopted in 1984, and

by the end of 1989 over 60% of large U.S. companies had adopted some form of poison pill (Davis, 1991).

Anecdotal evidence suggests that the interpreta- tions and attributions that our two groups made regarding poison pills were consistent with the the- oretically derived patterns described above. Within the investment community, the practice itself was portrayed in unequivocally negative terms. These reactions appeared to focus on poison pills’ in- terference with the market for corporate control (Walsh & Seward, 1990). Many investors voiced concerns that if underperforming managers could use such devices to sidestep the disciplining power of the market for corporate control, this would dra- matically weaken checks on managerial power (Jensen, 1988). Such accounts also tended to down- play the need to maintain corporate cohesiveness, consistent with the agency model’s conceptualiza- tion of the corporation as a changeable nexus of contracts. In line with the idea that negative actions by out-group members will be attributed to internal causes (Mackie & Smith, 1998; Pettigrew, 1979), those in the investment community often professed that poison pills were adopted primarily to preserve managers’ own positions and so revealed managers’ self-interested nature (Davis & Greve, 1997).

On the other hand, managers themselves often portrayed poison pills as a necessary defense against hostile takeovers and their associated undesirable effects on stakeholders. While managers tended to acknowledge they would rather not have to use them, they did label the practice in more positive terms (as “shark repellent”, for example) and argued that the practice maintained continuity that allowed executives and directors to meet their responsibili- ties to a broader array of non-shareholder groups, such as employees and communities (Hirsch, 1986; Useem, 1993). Top managers at times portrayed hos- tile acquirers in negative and emotionally charged terms, as seen in the justification offered by Michel C. Bergerac, Chairman and CEO of Revlon, for the firm’s adoption of a poison pill when faced with a hostile bid: “We are being met with poison so we are ap- plyingthenecessary antidotes” (Gilman & Hertzberg, 1985). This response is consistent with a view of governance wherein managers’ key task is to act for the good of the firm as a whole and balance the in- terests of different constituencies. It also provides some evidence regarding how managers could attri- bute the use of poison pills to external pressures rather than to negative managerial traits. These points suggest that peer-firm executives, when they observe other managers adopting poison pills, would

2015 1745Bednar, Love, and Kraatz

start with a less negative view of the practice itself. They also suggest that if peer-firm executives adopt an in-group perspective, as we have suggested they will, they will find ready explanations suggesting that external pressures caused adoption of the practice.

It is important to note that at best, poison pills were seen to protect firms from an undesirable negative outcome rather than being an outright positive action in their own right. Moreover, they were generally associated with negative stock price reactions (Ryngaert, 1988).2 Together, these points suggest that we are unlikely to see attributions of positive traits to managers implementing the practice, and we stipu- late that poison pills are likely to have a negative impact on managerial reputations in both audiences’ eyes. For analysts, however, the combination of a strongly negative interpretation accompanied by an internal attribution of self-serving character sug- gests that a substantial reputational penalty is likely. For peer executives, however, the combination of a more nuanced interpretation of the practice along with a likely external attribution toward situational pressures suggests a milder reputational penalty. Thus, we predict:

Hypothesis 1. Having a poison pill will do more damage to the reputation of the firm’s managers in the eyes of analysts than in the eyes of peer- firm executives.

Effects of Media Coverage

While we focus on how poison pills’ reputational consequences differ depending on the evaluating audience, another integral part of our argument is that those consequences may vary across firms and over time. We consider three specific sources of such variation, the first of which is media coverage of the practice. Media coverage is important be- cause it can affect whether managers’ actions come

to the attention of evaluators, as well as the meanings that evaluators attach to those actions (Fombrun, 1996; Pollock, Rindova, & Maggitti, 2008). Repu- tational evaluators must encode signals from an information-dense environment, and so not all reputational signals will come to their attention (Barnett, 2013; Ocasio, 1997). Moreover, journal- ists tend to construct dramatic narratives that at- tribute firm actions and outcomes to managers’ decision making, rather than to situational factors (Hayward, Rindova, & Pollock, 2004). In our par- ticular context, media accounts often portrayed poison pills negatively and, in some cases, even attributed their use to negative managerial traits, either directly or by quoting opponents of the practice. For example, according to an account in the Wall Street Journal, Union Carbide did not adopt a poison pill because it “smacked of mana- gerial entrenchment” (Stewart & Hertzberg, 1986). Other articles quoted shareholder activists who talked about poison pills as “designed to entrench existing management” and “as totally uncalled for and obviously against the interests of share- holders” (McCoy & Williams, 1985). To the extent that a poison pill is widely publicized in the media, then, reputational evaluators are (1) more likely to notice it and (2) more likely to make negative trait attributions about the managers using the practice, with the implication that evaluators will impose a larger reputational price than they otherwise would. By the same logic, a poison pill that does not receive media attention should inflict less reputational harm.

While these arguments apply generally, we also predict that media coverage will have less influence on the reputational penalties assessed by analysts than by executives. This proposed difference arises because analysts are likely to apply an out-group perspective while executives should view use of the practice using an in-group lens. We have already described how analysts’ status as out-group mem- bers leaves them inclined to make negative internal attributions when managers employ poison pills. Because of this, the negative attributions that appear in media reports seem likely to confirm, rather than change, analysts’ base interpretations. So, while we expect media reports to affect the reputational pen- alties that analysts impose on managers who adopt poison pills, we see this effect as primarily arising from media’s information-dissemination function, rather than from media-induced changes in how analysts interpret the practice and make attributions about managers who employ it.

2 While our focus is on what controversial gover- nance practices may be perceived to say about man- agers’ self-serving nature, we acknowledge that the reputational impact of these practices will be affected by their consequences for firm performance. We ac- count for such effects empirically by controlling for the impact of firm financial performance (including stock price performance) on managers’ reputations before we assess the reputational impact of poison pills. That is, we are interested in the reputational effects of poison pills over and above those emanating from their eco- nomic consequences.

1746 DecemberAcademy of Management Journal

Turning to executives, we have also already de- scribed how peer-firm evaluators will tend to look at other managers’ use of poison pills through an in- group lens, and so will tend to interpret use of poison pills as a response to external pressures. Media re- ports about specific firms and managers are unlikely to affect this general cognitive tendency, which is deeply rooted in group memberships. However, this same initial tendency means that executives do have plenty of room to move toward more internal attri- butions if they receive compelling information sup- porting such a change in specific situations. So, if executives encounter media reports that cast a par- ticular firm’s managers’ choice to use poison pills as a self-serving one, evaluations of those spe- cific managers’ competence and characters may be affected, rather than simply confirmed as with ana- lysts. This latter argument draws on social psycho- logical research, finding that some individuals in a group can come to be seen as distinctive, even while most are evaluated primarily as members of the group itself (Wilder & Thompson, 1980). In other words, even though fellow executives may generally attribute poison pill use to external factors, they could still be swayed to make internal attributions in specific cases, such as when there is media coverage of a particular firm’s use of poison pills. Overall, we expect executives to be more effected by media coverage not only because coverage makes them aware of poison pill usage, but also the interpreta- tions of these actions in media reports could sway executives’ own attributions about the managers using poison pills in those cases. Based on these arguments, we hypothesize:

Hypothesis 2a. Having a poison pill will do more damage to the reputation of the firm’s managers the more media coverage the poison pill receives.

Hypothesis2b. The effect outlined in Hypothesis 2a will be stronger for executives than for analysts.

Effects of Firm Performance

We also consider how the reputational penalty that managers bear for use of a poison pill is affected by their firm’s financial performance. A primary function of the market for corporate control is to enable owners to replace managers at poorly per- forming firms (Walsh & Seward, 1990). This implies that the firm’s financial performance is likely to be a salient contextual feature that may influence how

external evaluators make attributions about the use of poison pills.

That said, we suggest that financial performance will have different effects on evaluations by analysts compared to peer-firm executives. The difference arises because of how in-group versus out-group perspectives shape interpretations of poison pill use. If we consider analysts first, their out-group per- spective means that they are generally going to see use of a poison pill as reflecting managers’ self- protective tendencies, rather than as a response to situational pressures. If they apply this frame to a firm that is performing well, they should see the firm as less exposed to the market for corporate control, which should lead them to attribute its use to a free choice by the firm’s top managers. This in- creased tendency toward internal attribution would, in turn, strengthen the reputational penalty imposed for use of poison pill. Analysts, then, are likely to impose stronger penalties on managers at well- performing firms.

On the other hand, the influence of firm financial performance on executives’ attributions seems likely to run in the opposite direction. Executives’ in-group perspective means they are generally going to view adoption of a poison pill as a response to situational pressures, rather than to traits of the adopting man- agers. However, it is clear that managers have often been sensitive to concerns that poison pills could be adopted for self-protective reasons. So, if we com- pare executives’ evaluations of poison pill use at firms with strong and weak performance, we can see that under conditions of poor performance—where the firm is particularly exposed to the market for corporate control and thus managers’ jobs are more clearly at risk—that executives are more likely to consider the possibility of internal causal motiva- tions for use of poison pills. By this logic, when a poison pill is adopted at a firm that is performing well, executives are more likely to dismiss argu- ments that the executives were acting in self- protective ways because the immediate risk of losing one’s job would be lower. From the execu- tives’ perspective, use of a poison pill at a high- performing firm is not reactive but proactive. Executives may see fellow managers as operating from a position of strength to put a mechanism in place that will preserve the organization for the long term and protect it from shortsighted raiders. These arguments are supported by findings that people are able to hold both external and internal explanations for the same action (Morris & Peng, 1994), which is what we suggest happens here. Executives, then, are

2015 1747Bednar, Love, and Kraatz

likely to impose weaker penalties on managers at well-performing firms; thus:

Hypothesis 3a. When analysts are the evalua- tors, having a poison pill will do more damage to the reputation of the firm’s managers when the firm is performing well.

Hypothesis 3b. When executives are the evalu- ators, having a poison pill will do less damage to the reputation of the firm’s managers when the firm is performing well.

Effects of Prior Adoptions

The final contextual factor we consider is the ex- tent to which poison pills have been adopted by similar firms. This contingency is particularly rele- vant to our questions because it speaks to how rep- utational penalties associated with controversial governance practices may be sustained (or not) as the practice spreads. The contingency has also played a prominent role in much work in institutional the- ory and elsewhere (Rogers, 1995), with such work generally arguing that observers’ interpretations of a practice become more positive as it becomes more widely adopted, even if it is initially considered controversial or deviant. Specifically, the practice gains in standing because observers see widespread use as evidence of the practice’s appropriateness (Abrahamson & Rosenkopf, 1993; Tolbert & Zucker, 1983). However, poison pills, like many other con- troversial governance practices, do not appear to completely conform to this pattern because they continue to be viewed quite negatively and attract controversy, even though they have been in wide- spread use for more than two decades (Velasco, 2001). If we stay within the institution-theoretic frame, which focuses on perceptions of the practice itself, these persistent negative interpretations might lead us to expect that reputational penalties associ- ated with poison pills’ use will show little change as it spreads.

We propose, however, that even though views of the practice itself remain relatively negative, widespread use is likely to change causal attribu- tions regarding the managers involved. Causal at- tributions regarding a focal actors’ behavior depend on those of similar others (Kelley, 1967). If few ac- tors engage in a behavior, outside parties tend to attribute it to internal, dispositional factors. But if many actors are engaged in the same behavior, at- tributions to common external factors become more likely (Kelley, 1967). Applying this to our context,

we suggest that as poison pills become more com- mon, evaluators will be more prone to making external attributions, and so the reputational pen- alty that managers bear should diminish. This logic draws strength from its consistency with the “safety in numbers” argument advanced in other research on controversial practices (Ahmadjian & Robinson, 2001).

While this proposed effect is not dependent on the evaluating audience, we posit that it should in- fluence analysts more than executives. We have al- ready argued that as out-group members, analysts will be biased toward making internal causal attri- butions when executives use poison pills, and that they will generally tend to impose larger reputa- tional penalties compared to executives. At the same time, analysts should still be affected by the ten- dency to make external attributions when many actors engage in similar behaviors, as social com- parisons deeply affect attributions of actors (Kelley, 1973). This suggests that the reputational penalty that analysts impose may substantially diminish as poison pills spread. On the other hand, we have also argued that executives, as in-group members, will be biased from the start toward making external causal attributions for executives’ use of poison pills. Ex- ecutives seem unlikely to make major changes in their attributions as poison pills become wide- spread, because their initial views are already largely congruent with the external attributions that widespread use tends to elicit. So, we expect not only that the reputational penalty that executives impose for poison pill use will be relatively small overall (as in Hypothesis 1), but the penalty will also not change very much as poison pill use becomes widespread. Thus:

Hypothesis 4. Having a poison pill will do less damage to the reputation of the firm’s managers as more poison pills are adopted.

Hypothesis 4a. The effect outlined in Hypothesis 4a will be stronger for analysts than for executives.

METHODS

Data and Sample

Our sample was drawn from the largest U.S. in- dustrial firms during the 1980s. We followed Davis (1991) in including publicly traded firms that appeared on either the 1985 or the 1980 Fortune 500 list of the largest U.S. firms by sales. This

1748 DecemberAcademy of Management Journal

sample is an attractive one because these firms represented the core of American industrial firms during the study period, they adopted poison pills widely, and they have been the subject of prior studies of the practice (e.g., Davis, 1991; Davis & Greve, 1997). In addition, Fortune “Most Admired Companies” data are required for firms to be in the final sample, and are available only for such large firms.

Our study period was 1985 to 1989. The initial year is early in the poison pills’ history, as the first adoption by a Fortune 500 firm occurred in August 1984. The study period starts in 1985 because dis- aggregated Fortune survey data were unavailable in prior years. The period ends in 1989 because con- tinuous data on poison pill usage were not available to us after that year and the practice had come into widespread use by that time (Davis, 1991). Data on poison pills, board composition, and other anti- takeover provisions were generously provided by Jerry Davis, who collected them from the Investor Responsibility Research Center dataset. Financial performance data were from COMPUSTAT, and media visibility data were collected from articles in the Wall Street Journal. Control variables were taken from a variety of sources, including media reports in the Wall Street Journal and corporate manuals found through Mergent WebReports. The final sample in- cludes 266 firms and the 1,108 firm-years for which all key data were available.

Measures

Dependent variables. In order to test our hy- potheses, we needed measures of managerial rep- utation as assessed by different audiences. We obtained these from the Fortune survey of Amer- ica’s “Most Admired Companies.” Since the early 1980s, Fortune has annually polled thousands of executives (including board members) and stock analysts, asking them to rate the overall quality of the 10 largest firms in their industry by rating each firm along eight different dimensions. One of these dimensions is management quality, which we used to create two dependent measures of managerial reputation. Managerial Reputation—By Executives is the score on the “managerial quality” item as rated by executive respondents, while Managerial Reputation—By Analysts is the same measure as rated by analysts. This measure is con- sistent with our conception of managerial reputa- tion as an overall, perceptual evaluation of the quality of managers’ role performance, in the eyes of

a specific audience. Some scholars have suggested that removing the performance component of For- tune reputations enables more effective assessment of the reputational impact of other types of signals (Brown & Perry, 1994; Fombrun & Shanley, 1990). Consequently, we transformed our two dependent variables by regressing each on several measures of firm performance (log of sales, profitability, change in sales, change in market value, and lagged change in sales and market value), and then using the regressions’ residuals as the actual dependent variables in the analyses reported below.3 This transformation also accounts for the reputational impact associated with poison pills’ performance consequences (e.g., stock price declines), with the implication that any effects we do observe from them arise from factors other than direct economic ones.

Our hypotheses also predict differences in the strength of the reputational penalty associated with poison pills depending on the evaluating audience. To assess such effects, we created a measure of the difference in reputational evaluations between au- diences by taking the managerial reputation item score for executives and subtracting the score for analysts. To the extent that analysts and executives have significant differences in their evaluations of managers, we should see the regression coefficients of our independent variables and hypothesized in- teractions to be significantly related to this variable. For example, if analysts judge managers with poison pills more harshly than executives, we would expect the coefficient for poison pill to be positive and significant.

Independent Variables. Our core independent variable, poison pill, is a dichotomous, annual measure set to one if a firm had a poison pill an- nounced or in force during the relevant year.4 For

3 We also estimated models in which we used un- transformed dependent variables and included the per- formance measures as controls. The results were similar to those reported here.

4 We took the timing of the Fortune survey itself into account in structuring our analyses. This survey is mailed out late in the calendar year (October or November), and results are published early the next year. Thus, we con- sidered a firm to have a poison pill for a given year if the pill was in force or had been publicly reported by the end of November of that year. Financial-performance measures were similarly constructed using the four quarters prior to the survey process, rather than using calendar-year or fiscal-year boundaries.

2015 1749Bednar, Love, and Kraatz

example, if a firm adopted a poison pill in June 1987, this variable would be zero for 1985 and 1986, and one for 1987, 1988, and 1989 (assuming that the poison pill remained in force during the later years, as would be typical). This measure is consistent with our conceptualization of poison pills as an ongoing governance arrangement that amounts to a characteristic of the firm and/or its managers.

We measured media coverage of poison pill as the number of articles in which a firm was mentioned along with the term “poison pill” in a given year in the Wall Street Journal. A trained coder read each article to ensure that the poison pill reference was actually about the focal firm.

To test our hypotheses about the moderating effect of firm performance, we created the in- teraction term poison pill * profitability, where profitability is the return on book assets in the relevant year. We centered this variable by sub- tracting its mean value across the full sample. While centering does not substantively affect the interaction term’s significance, it does facilitate interpretation of the findings (Jaccard, Turrisi, & Wan, 1990).

Finally, we measured the extent of adoption of poison pills using the variable prior adoptions, which is the cumulative total of firms in our sam- ple that had adopted a poison pill. The interac- tion term poison pill * prior adoptions assesses differences in poison pills’ reputational impact as the practice spread.5 We centered the variable prior adoptions to facilitate interpretation of the interaction term.

Control Variables. We included controls for several prominent governance characteristics. CEO duality was coded as a dummy variable equal to one if the CEO also held the board chair position in a given year. We controlled for CEO tenure, since long tenure may indicate managerial entrench- ment (Berger, Ofek, & Yermack, 1997; Finkelstein & Hambrick, 1990). Board independence (or lack thereof) was measured as the percentage of insiders who were members of the board in the year 1986. The level of institutional ownership was measured as the percentage of total shares owned by in- stitutions during the year 1986. Institutional own- ership has been important in governance research

and is also positively related to corporate reputa- tion (Fombrun & Shanley, 1990). We used a single year of data for the latter two measures because they are quite stable over time and because of the diffi- culty in collecting these data beyond the single year.6

We included several other controls. Performance controls are built into our models because the de- pendent variables are residuals from which the performance component has been removed. We did, however, include a measure for firm profit- ability as a base term for the poison pill * profit- ability measure. To avoid conflating the impact of takeover threats with those of our anti-takeover measure, we created a measure of firm takeover risk by searching the Wall Street Journal for articles wherein the focal firm was mentioned along with the word “takeover.” A team of two coders read the resulting articles and created a dichoto- mous variable equal to “1” if the firm was described as being at risk of a takeover in the focal year. We also included year and industry dummies in the models.

One potential concern arises because unob- served factors could influence both the likeli- hood of having a poison pill and the reputation of the firm and its managers. To account for this potential endogeneity, we follow prior studies (Chatterjee & Hambrick, 2007; Pollock & Rindova, 2003; Sanders & Hambrick, 2007) by using a first- stage regression to predict the likelihood of the potentially problematic event, and then control for that predicted likelihood by including the in- verse mills ratio in the main regression (Heckman, 1979; Shaver, 1998). We used a probit model to estimate the likelihood of a focal firm having a poison pill based on the firm’s use of other anti- takeover provisions, as such provisions are usu- ally not adopted in isolation (results available

5 We tested for curvilinear effects of prior adoption, but the interaction between poison pills and the square of prior adoptions was not significant in any of the models.

6 We did have 1982 board membership data from another project, and found that the percentage of in- siders was correlated to .693 in that year with the 1986 board data. This supported our belief about the stability of the insider percentage and the validity of using a single year of board membership data. In addition, we were unable to collect institutional ownership data at all in 13% of our cases. To avoid losing these observa- tions, we used the average level of ownership across the sample where values were missing. In other models, we created a dummy variable equal to “1” in cases where this variable was missing and the results were sub- stantively similar.

1750 DecemberAcademy of Management Journal

from authors on request).7 The predicted likeli- hood from the probit regression was transformed into an inverse mills ratio, which was then in- cluded in the main models.

Analysis

We report analyses that use generalized-least-square (GLS) cross-sectional time series regression (xtgls in STATA), modeling first-order autocorrelation in the error terms and correcting for potential hetero- skedasticity. Cross-sectional time-seriesregression is appropriate for panel data analysis with a continuous dependent variable such as that examined here (Hsiao, 1986). GLS models are also advantageous because they allow for autocorrelation in error terms.

This is ideal for a dependent variable such as repu- tation, where persistent firm-level effects clearly exist but those effects may also change over time (for ex- ample, IBM was perceived as a model of the well- managed firm in the 1980s but faltered and was dismissed as an unresponsive bureaucracy in the early 1990s). Indeed, the estimated year-to-year error correlation was typically between .4 and .7, as shown in theanalyses presented below. This is far too high to ignore, but also too low to justify assuming stable firm-specific effects over the five-year study period.

RESULTS

Table 1 displays descriptive statistics and a corre- lation matrix for all variables. For the profitability and prior-adoptions measures, we present the means prior to centering.

Tables 2 and 3 present our analyses of poison pills’ effects on managerial reputation in the eyes of peer executives and analysts. Models 1a and 1b in Table 2 are baseline models for executives and analysts, re- spectively. Models 2a and 2b add the main effect of poison pills, while Models 3a and 3b show the effect of the media coverage variable. In Table 3, Models 4a and 4b show the interaction effects for profitability, Models 5a and 5b show the interaction of prior adop- tions, and Models 6a and 6b are the full models.

TABLE 1 Correlation Table and Descriptive Statisticsa

Mean SD 1 2 3 4 5 6

1 Managerial reputation—by analysts 0.001 1.031 1.000 2 Managerial reputation—by executives 0.005 0.858 0.770 1.000 3 Managerial reputation—difference 0.004 0.661 20.560 0.097 1.000 4 Poison pill 0.355 0.479 20.099 20.138 20.025 1.000 5 Media coverage of poison pill 0.100 0.580 20.125 20.128 0.029 0.233 1.000 6 Profitability 5.451 5.090 0.112 0.163 0.038 20.048 20.061 1.000 7 Prior adoptions 78.996 41.041 0.056 20.003 20.091 0.391 0.062 20.039 8 CEO tenure 7.734 6.923 0.037 0.003 20.055 20.006 20.068 0.099 9 CEO duality 0.782 0.413 20.025 20.024 0.007 0.126 20.037 0.002 10 Insider ratio 27.705 13.519 0.050 0.113 0.069 20.108 20.014 0.146 11 Institutional ownership 50.003 14.332 0.115 0.134 20.006 0.208 0.029 0.015 12 Takeover threat 0.008 0.090 20.051 20.058 0.004 0.038 0.384 0.017 13 Inverse mills ratio 6.952 8.805 20.074 20.034 0.072 20.483 20.105 0.067

7 8 9 10 11 12 13 7 Prior adoptions 1.000 8 CEO tenure 0.023 1.000 9 CEO duality 0.062 0.068 1.000 10 Insider ratio 20.009 0.027 20.106 1.000 11 Institutional ownership 20.030 20.035 0.076 20.120 1.000 12 Takeover Threat 20.009 20.004 20.050 0.016 20.015 1.000 13 Inverse Mills Ratio 20.785 0.035 20.188 0.178 20.232 20.021 1.000

a n 5 1,107 firm years.

7 The other anti-takeover provisions were classified boards and unequal voting rights. These were appropriate for our purposes, because they were correlated with the use of poison pills, but not with managerial reputation. In our probit model, classified boards (p , .001) were posi- tively and significantly related to the likelihood of having a poison pill. On the other hand, unequal voting rights were negatively related to poison pill use (p , .005). As with our measure of institutional ownership, we were only able to access these data for approximately 75% of the firms in our sample, so we used the average across the sample for cases where these data were missing.

2015 1751Bednar, Love, and Kraatz

Model 7 is specified similarly to Models 6a and 6b, but tests for significant differences between execu- tives and analysts by using the difference between executive and analyst reputational evaluations as the dependent variable. We interpret coefficients for Hypothesis 1 using Models 2a and 2b, and the remain- ing hypotheses using the full models (6a, 6b and 7).

The models reveal support for Hypothesis 1’s prediction that poison pills would damage manage- rial reputation more in the eyesof analysts thanin the eyes of executives. The coefficient for poison pill in Model 2a, indicating the average predicted reputa- tional effect across our sample, was marginally sig- nificant when peer executives were the evaluating audience (b 5 2 .053, p , .10), but when analysts were the evaluators the reputational penalty was highly significant, and almost three times as severe (Model 2b, b 5 2 .153, p , .001). Model 7 provides evidence that this difference was also statistically significant (p , .01). To provide a sense of the effect

size, consider that Fortune ranks firms in each in- dustry according to their reputation ratings, and the mean rating difference between consecutively ranked firms is approximately .32 points. In the eyes of analysts, poison pills thus appear to cost a firm approximately half of a ranking in the Fortune survey—a substantial effect when considered in light of the fact poison pills are just one governance practice and that the reputation measures are overall assessments of the firm’s management team.

Hypothesis 2a, that media coverage of a firm’s poison pill would amplify the size of the reputational penalty associated with that poison pill, receives partial support. Specifically, media coverage had a significant negative impact on managerial repu- tation in the eyes of executives (Model 6a, b 5 .045, p , .01), but not in the eyes of analysts (Model 6b, b 5 .058, n.s.). However, there is no support for Hypothesis 2b’s prediction that this effect would be greater for executives than for analysts as Model 7,

TABLE 2 Influence of Poison Pills on Managerial Reputationa

Executives Analysts Executives Analysts Executives Analysts

1a 1b 2a 2b 3a 3b Profitability 0.006* 0.0081 0.000 0.011* 0.001 0.011*

(0.003) (0.005) (0.003) (0.005) (0.003) (0.005) CEO tenure 20.001 0.001 0.000 0.000 20.001 20.001

(0.002) (0.004) (0.002) (0.003) (0.002) (0.003) CEO duality 20.110** 20.061 20.114** 20.043 20.114** 20.048

(0.042) (0.057) (0.042) (0.054) (0.042) (0.054) Insider ratio 0.008*** 0.006* 0.008*** 0.006** 0.008*** 0.006*

(0.002) (0.002) (0.002) (0.002) (0.002) (0.002) Inst. ownership 0.011*** 0.011*** 0.011*** 0.012*** 0.011*** 0.012***

(0.002) (0.003) (0.002) (0.003) (0.002) (0.003) Threat of takeover 20.358** 20.144 20.342*** 20.157 20.240* 20.036

(0.111) (0.167) (0.099) (0.173) (0.111) (0.197) Inverse mills ratio 20.013* 20.001 20.011* 20.004 20.012* 20.003

(0.006) (0.008) (0.006) (0.008) (0.006) (0.008) Prior adoptions 20.003*** 0.000 20.003** 0.001 20.003** 0.001

(0.001) (0.001) (0.001) (0.001) (0.001) (0.001) 1986 0.032 0.024 0.042 0.029 0.037 0.030

(0.036) (0.052) (0.036) (0.053) (0.036) (0.053) 1987 0.051 0.143** 0.043 0.147** 0.040 0.143**

(0.033) (0.048) (0.032) (0.048) (0.033) (0.048) 1988 0.045* 0.048 0.0381 0.0551 0.0351 0.050

(0.022) (0.032) (0.021) (0.033) (0.021) (0.033) Poison pill 20.0531 20.153*** 20.0551 20.160***

(0.031) (0.046) (0.031) (0.046) Media coverage of Poison pill 20.047* (0.022) 20.054 (0.039) Poison pill 3 Profitability Poison pill 3 prior adoptions Firms 266 266 266 266 266 266 Firm years 1103 1103 1103 1103 1103 1103

1p , 0.10, * p , 0.05, ** p , 0.01, *** p , 0.001, two-tailed tests a Year and industry dummies included.

1752 DecemberAcademy of Management Journal

T A B L E 3

In fl u en

ce o f P o is o n P il ls

o n M a n a ge

ri a l R ep

u ta ti o n (c o n t’ d ) a

E x ec u ti v es

A n a ly st s

E x ec u ti v es

A n a ly st s

E x ec u ti v es

A n a ly st s

D if fe re n ce

4 a

4 b

5 a

5 b

6 a

6 b

7 P ro fi ta b il it y

2 0 .0 0 1

0 .0 1 7 * *

2 0 .0 0 1

0 .0 1 1 *

2 0 .0 0 1

0 .0 1 8 * *

2 0 .0 1 1 * *

(0 .0 0 3 )

(0 .0 0 6 )

(0 .0 0 3 )

(0 .0 0 5 )

(0 .0 0 3 )

(0 .0 0 6 )

(0 .0 0 4 )

C E O

te n u re

0 .0 0 0

2 0 .0 0 1

0 .0 0 0

2 0 .0 0 1

2 0 .0 0 1

2 0 .0 0 1

2 0 .0 0 1

(0 .0 0 2 )

(0 .0 0 3 )

(0 .0 0 2 )

(0 .0 0 3 )

(0 .0 0 2 )

(0 .0 0 3 )

(0 .0 0 2 )

C E O

d u al it y

2 0 .1 1 3 * *

2 0 .0 3 8

2 0 .1 1 5 * *

2 0 .0 5 2

2 0 .1 1 2 * *

2 0 .0 5 5

2 0 .0 0 6

(0 .0 4 2 )

(0 .0 5 4 )

(0 .0 4 2 )

(0 .0 5 5 )

(0 .0 4 2 )

(0 .0 5 5 )

(0 .0 3 7 )

In si d er

ra ti o

0 .0 0 8 * * *

0 .0 0 6 *

0 .0 0 8 * * *

0 .0 0 6 * *

0 .0 0 8 * * *

0 .0 0 6 *

0 .0 0 3 * *

(0 .0 0 2 )

(0 .0 0 2 )

(0 .0 0 2 )

(0 .0 0 2 )

(0 .0 0 2 )

(0 .0 0 2 )

(0 .0 0 1 )

In st .O

w n er sh

ip 0 .0 1 1 * * *

0 .0 1 2 * * *

0 .0 1 1 * * *

0 .0 1 2 * * *

0 .0 1 1 * * *

0 .0 1 2 * * *

0 .0 0 2

(0 .0 0 2 )

(0 .0 0 3 )

(0 .0 0 2 )

(0 .0 0 3 )

(0 .0 0 2 )

(0 .0 0 3 )

(0 .0 0 1 )

T h re at

o f ta k eo

v er

2 0 .3 4 3 * * *

2 0 .1 4 3

2 0 .3 4 1 * * *

2 0 .1 5 2

2 0 .2 4 5 *

0 .0 2 5

2 0 .1 6 1

(0 .0 9 9 )

(0 .1 6 8 )

(0 .0 9 6 )

(0 .1 7 4 )

(0 .1 0 9 )

(0 .1 9 6 )

(0 .1 7 4 )

In v er se

M il ls

R at io

2 0 .0 1 1 1

2 0 .0 0 3

2 0 .0 1 1 1

2 0 .0 0 5

2 0 .0 1 1 1

2 0 .0 0 3

2 0 .0 0 5

(0 .0 0 6 )

(0 .0 0 8 )

(0 .0 0 6 )

(0 .0 0 8 )

(0 .0 0 6 )

(0 .0 0 8 )

(0 .0 0 5 )

P ri o r ad

o p ti o n s

2 0 .0 0 3 * *

0 .0 0 1

2 0 .0 0 3 * *

0 .0 0 0

2 0 .0 0 3 * *

0 .0 0 0

2 0 .0 0 2 1

(0 .0 0 1 )

(0 .0 0 1 )

(0 .0 0 1 )

(0 .0 0 1 )

(0 .0 0 1 )

(0 .0 0 1 )

(0 .0 0 1 )

1 9 8 6

0 .0 4 5

0 .0 3 3

0 .0 4 2

0 .0 6 8

0 .0 3 5

0 .0 7 9

2 0 .0 7 9 *

(0 .0 3 6 )

(0 .0 5 3 )

(0 .0 3 8 )

(0 .0 5 6 )

(0 .0 3 9 )

(0 .0 5 7 )

(0 .0 4 0 )

1 9 8 7

0 .0 4 6

0 .1 5 0 * *

0 .0 4 0

0 .1 8 2 * * *

0 .0 3 6

0 .1 8 6 * * *

2 0 .1 9 6 * * *

(0 .0 3 2 )

(0 .0 4 8 )

(0 .0 3 4 )

(0 .0 5 1 )

(0 .0 3 5 )

(0 .0 5 1 )

(0 .0 3 9 )

1 9 8 8

0 .0 4 0 1

0 .0 4 8

0 .0 3 6 1

0 .0 7 1 *

0 .0 3 4

0 .0 5 9 1

2 0 .0 3 7

(0 .0 2 1 )

(0 .0 3 3 )

(0 .0 2 1 )

(0 .0 3 4 )

(0 .0 2 2 )

(0 .0 3 4 )

(0 .0 2 9 )

P o is o n P il l

2 0 .0 5 3 1

2 0 .1 4 0 * *

2 0 .0 4 8

2 0 .1 9 2 * * *

2 0 .0 4 9

2 0 .1 9 1 * * *

0 .1 0 9 * *

(0 .0 3 1 )

(0 .0 4 7 )

(0 .0 3 1 )

(0 .0 4 9 )

(0 .0 3 2 )

(0 .0 5 1 )

(0 .0 4 1 )

M ed

ia C o v er a ge

o f

P o is o n P il l

2 0 .0 4 5 * (0 .0 2 1 )

2 0 .0 5 8 (0 .0 3 8 )

0 .0 2 3 (0 .0 2 5 )

P o is o n P il l 3

P ro fi ta b il it y

0 .0 0 4 (0 .0 0 4 )

2 0 .0 1 9 * (0 .0 0 8 )

0 .0 0 3 (0 .0 0 4 )

2 0 .0 2 1 * * (0 .0 0 8 )

0 .0 2 3 * * * (0 .0 0 6 )

P o is o n P il l 3

P ri o r

A d o p ti o n s

0 .0 0 0 (0 .0 0 1 )

0 .0 0 3 1 (0 .0 0 2 )

0 .0 0 0 (0 .0 0 1 )

0 .0 0 3 * (0 .0 0 2 )

2 0 .0 0 4 * * (0 .0 0 1 )

F ir m s

2 6 6

2 6 6

2 6 6

2 6 6

2 6 6

2 6 6

2 6 6

F ir m

Y ea

rs 1 1 0 3

1 1 0 3

1 1 0 3

1 1 0 3

1 1 0 3

1 1 0 3

1 1 0 3

1 p ,

.1 0 ,*

p ,

.0 5 ,*

* p ,

.0 1 ,*

* * p ,

.0 0 1 ,t w o -t ai le d te st s

a Y ea

r an

d in d u st ry

d u m m ie s in cl u d ed

.

2015 1753Bednar, Love, and Kraatz

the difference model, shows that the coefficient for media coverage was not statistically significant.

Hypothesis 3a predicts a negative interaction ef- fect between profitability and the presence of a poi- son pill when analysts are the evaluating audience, while Hypothesis 3b predicts that this effect will be reversed for executives. The coefficient for the interaction term poison pill * profitability is nega- tive and significant for analysts (Model 6b, b 5 2.021, p , .01), supporting Hypothesis 3a. Hypoth- esis 3b, however, is not supported as the coeffi- cient is positive (as predicted) but not significant (Model 6a, b 5 .003, n.s.) That said, the difference between the two coefficients is significant, as shown in Model 7, indicating that the effect of firm perfor- mance is different for the two audiences even if that difference is not as dramatic (i.e., reversed effects), as predicted. Figure 1 displays the reputational penalty associated with having a poison pill for both execu- tives and analysts (note that the penalty increases as one moves down on the figure). At levels of perfor- mance one standard deviation below the mean, the effect of having a poison pill is small and quite similar for both executives and analysts. However, as shown by the widening distancebetween the two lineson the graph, the reputational penalty grows in the eyes of analysts as performance approaches one standard deviation above the mean, while remaining relatively unchanged for executives. At performance levels one standard deviation above the mean, the reputational penalty imposed by analysts is more than six times larger than the penalty imposed by executives.

Hypothesis 4a predicts that as poison pills diffuse, the reputational penalty associated with their use will decrease, while Hypothesis 4b predicts that this effect will be more pronounced for analysts than executives. Both hypotheses receive support. For analysts, the interaction term poison pill * prior

adoptionis positive and significant (b 5 .003, p , .05, Model 6a) but it is not statistically significant for ex- ecutives (b 5 .000, n.s., Model 6b). As shown in Model 7, the difference in the coefficients for these two groups is itself statistically significant. Figure 2 shows how the marginal penalty associated with use of a poison pill changed as more firms adopted the practice. There is a wide disparity between analysts and executives when relatively few others had adopted the practice. Indeed, for the initial adopters in our sample, the reputational damage in the eyes of analysts was the equivalent of managers’ losing ap- proximately a rank and a half in the Fortune survey. However, as shown by the downward-sloping line for analysts, this substantial reputational penalty quickly dissipated as more firms adopted poison pills. For executives, the penalty remains small and relatively stable. The reputational evaluations of an- alystsandexecutivesnearly convergeatthetopofour observed range of prior adoptions (which is 121 firms) at a modest level of reputational penalty, al- though the actual point of convergence is outside the observable range of our data.8

DISCUSSION

Our general motivation in undertaking this study was to better understand the reputational penalties

FIGURE 1 Effect of Poison Pill on Managerial Reputation as Performance Increases

8 In supplementary analysis, we also ran our models using corporate reputation as the dependent measure, which consists of the average Fortune survey score for managerial quality plus seven additional items ranging from value as an investment to innovativeness. The results were consistently weaker in these models, which supports our assumption that controversial governance practices such as poison pills should be especially impactful for managerial reputation.

1754 DecemberAcademy of Management Journal

for questionable behaviors, which act as the foun- dation of reputation’s ability to serve a social control function. To do this, we developed and tested theory about the reputational price that managers are likely to pay for use of controversial governance practices, focusing particularly on poison pills in the empiri- cal study. We argued that existing theoretical treat- ments of reputation’s social function in governance research have been somewhat oversimplified and undersocialized. To begin to address these issues, we developed theory that accounts for the presence of multiple evaluating audiences and allows for dif- ferences in the causal attributions these groups make when assessing managerial reputation. Our findings support most of our conceptual arguments and make contributions to both the governance and reputation literatures. They also bring these two bodies of work together to better explain how and when reputation can act as a social control mechanism.

Our study’s contribution to the governance litera- ture starts with its empirical examination of the basic assumption in agency theory and elsewhere that managers’ reputations will suffer when they use governance practices that owners or other keyparties disapprove of. Our study reveals that although this assumption holds in some cases, in many cases it does not. What emerges from our findings is a highly contingent view of reputational penalties that ensue from adopting a controversial governance practice like poison pills. We found that the severity of the penalty varied greatly depending on which audience (analysts or executives) evaluated managerial repu- tations. The reputational damage also varied accord- ing to the situation. For instance, poison pills were associated with quite large reputational penalties in the eyes of analysts when the media covered the poi- son pill extensively, when the firm was performing well and when few other firms had adopted a poison pill. But when these circumstances were reversed, the

reputational penalty associated with poison pills could weaken greatly. The findings indicate that reputational penalties are real, and that they are at times meaningful, but that to understand the extent to which reputation may function as a social control device we need to account for contingencies that are likely to strengthen and weaken these penalties.

Our study’s examination of differences in audi- ences’ reputational penalties extends existing gov- ernance work. This work tends to be concerned with the views of a single evaluating audience (i.e., shareholders), but the findings indicate that important audiences diverged in their reputational reactions to poison pills. While it was beyond our scope to examine the consequences of these audi- ence differences for managerial decision making, one possible implication is that their fragmented reputational reactions may have substantially un- dercut reputation’s potential ability to deter the spread of poison pills. Managers may have taken heart in the relatively neutral reputational judgments of their peers, and thus placed less weight on the strong initial negative reactions of the investment community than they might have otherwise. While necessarily speculative, this argument is broadly consistent with recent ideas about pluralism in in- stitutional theory (Kraatz & Block, 2008), as well as those in research on poison pills’ diffusion and re- actions to them (Davis, 1991; Hirsch, 1986). It also suggests that differences in audiences’ reputational reactions may be an important factor underlying the widespread presence of controversial governance behaviors more generally. Further research that ex- plores how differences in audiences’ reputational judgments can influence managers’ decisions would thus be particularly valuable.

We also emphasized the role of causal attributions in shaping audiences’ reputational assessments, and particularly focused on how attributions are shaped

FIGURE 2 Effect of Poison Pill on Managerial Reputation as Prior Adoptions Increase

2015 1755Bednar, Love, and Kraatz

by in-group and out-group processes. We found ev- idence that these factors can lead to significant dif- ferences in the reputational penalties managers incur for use of poison pills, both in general and in terms of how key situational contingencies influence those effects. These arguments and findings signifi- cantly extend existing work on differences between reputation-granting audiences. This research has applied either economic or institutional lenses, and has focused almost exclusively on how different audiences directly assess a practice, behavior, or outcome. Our insight is that the reputational impact of these cues can also be importantly influenced by the attributions that they lead evaluators to make of the actors involved, and that those attributions may also be influenced by group membership, identity, and social structure more generally.

One example of this is seen in our rationale for Hypothesis 3a about the effect of firm performance on the reputational penalties applied by analysts. We applied attributional and in-group and out-group arguments to propose that these evaluators would apply a steeper penalty when firm performance was strong, and found support for that prediction. In contrast, a strictly economic view would be more likely to predict that poison pills produce a more severe reputational penalty (in analysts’ eyes) when the firm is performing poorly, because interference with the market for corporate control and damage to investors’ interests is more likely in such a situation. A second example arises in the case of prior adop- tions’ moderating effects. If we focus on views of the practices’ legitimacy or social standing alone, rather than considering attributions likely to be made about its use, we might expect a sustained negative reputa- tional reaction to poison pills. As mentioned above, poison pills remain controversial today and still at- tract charges of managerial self-interest. However, despite what appears to be continued contestation regarding poison pills’ legitimacy (Velasco, 2001), we found that the substantial reputational penalty ini- tially imposed by analysts dissipated dramatically as more firms implemented the practice. Our theory explains this result in large part as a consequence of changing causal attributions and in-group versus out- group differences. These examples highlight the po- tentially powerful effects, adding causal attributions and group membershipasexplanatoryfactorstothose that have been emphasized in prior work.

Finally, the media appears to play a significant role in determining the strength of the reputational pen- alty, which has not been adequately recognized in existing treatments of reputation as social control.

Our findings are consistent with the idea of the media serving as a corporate governance mechanism by bringing to light otherwise unknown actions, and strengthening the reputational sanctions for practices that are considered bad governance (Bednar, 2012; Bednar, Boivie, & Prince,2013; Dyck & Zingales,2002; King & Soule, 2007). It also supports the idea that journalists tend to attribute firm actions to individual managers (Hayward et al., 2004). When it comes to controversial governancepractices, themediaappears to play an important role as social arbiter in assessing the merits of, and passing judgments on, a particular practice, and the managers associated with its use (Wiesenfeld, Wurthmann, & Hambrick, 2008). The practical importance of media coverage is seen in the finding that when it was absent the penalty was much reduced, at least in the eyes of peer executives. This suggests that in many circumstances, managers may be able to “fly under the radar” with respect to their firms’ controversial governance practices.

Our findings also have implications for research on reputation as social control more broadly, beyond the governance context. Because prior research shows that poison pills diffused rapidly and broadly (Davis, 1991) we know that reputation broke down as a social control mechanism with respect to stopping the spread of this practice. By measuring the actual reputational penalty associated with poison pills’ use, we now know more about a key factor associated with this breakdown. One potential broader im- plication is that classic conceptualizations of how reputation serves a social control function in the reputation literature may suffer from some of the same limitations as those in the governance litera- ture addressed above. For instance, the reputa- tion literature has recognized that reputation is an audience-specific evaluation (Kim & Jensen, 2014), but this idea is typically overlooked in conceptuali- zations of reputation as social control. Differences between groups are likely to arise from different in- terpretations of a practice, but also from different causal attributions. Our findings suggest that actors and evaluators should be particularly aware of which groups are considered peers, since peer eval- uators may take into account situational factors in the face of negative signals. Our results also point to a dramatic decline in reputational penalty once the first few actors have taken a controversial action. This suggests that the social control function of reputation is difficult to maintain in the face of defections. The findings also suggest that broadcast- ing information about potentially negative actions strengthens the accompanying reputational penalty.

1756 DecemberAcademy of Management Journal

In sum, we do not dispute that reputation does play a social control function, but the findings here suggest it may be less robust, more contingent and harder to sustain than previously thought.

Like all research, our study has some limitations that point toward rich areas for future research. For one, although we motivated our study with reference to reputation’s social control function, our study focused on a single though critical element in this mechanism: reputational penalties. Future research could explore other elements, such as how the threat ofreputationallossactuallytranslatesintoamanager’s future actions. If we apply a parallel logic to that in our study, future research could consider how this trans- lation occurs in the presence of different evaluating audiences, and how it varies across firms and over time. Another limitation and opportunity arises from the critical role that causal attributions may play in reputation’s ability to deter questionable behaviors. While we presented empirical findings consistent with such a role and found anecdotal evidence in media accounts, our data provides no way to directly measure causal attributions. Exploring different eval- uators’ attributions regarding questionable behaviors seems a particularly rich area for future research.

There are also some potential boundary conditions that apply to our study. We focused on a specific time period when poison pills were extremely controver- sial, and found that the differences between audiences tended to dissipate over time. Thus, it is not clear that our results would persist in a later time period. We should also note that while other studies of the social control function of reputation have looked at closed communities, our sample consisted of a relatively open community. This may help to explain why we didnotseemoreseverereputationaldeclines,andwhy poison pills spread widely in spite of the reputational impacts that we did observe. Future research could further explore both the temporal and social-structural limits of reputation’s social control function.

Finally, we focused on just one practice within the theoretical category of controversial governance practices. Our results have particular implications for some practices that are controversial but long- standing, such as CEO duality, option re-pricing, and various other governance arrangements. It may be that if these practices are widely adopted, their capacity to generate reputational losses has dissi- pated even though they continue to attract disap- proval. Further investigation into the assumption that questionable behavior leads to reputational penalties is warranted, given that our findings sug- gest this relationship cannot be taken for granted.

Stepping back, it seems remarkable that a sin- gle governance practice could affect an important stakeholders’ overall evaluation of management enough to potentially cause changes in the firms’ standing in their industries, and yet managers still adopted that practice in great numbers. It is also re- markable that this penalty dissipated to the point of likely irrelevance within a few years, even while the behavior itself continued to attract controversy and negative attributions in the press. We believe these findings show the promise of research that illumi- nates reputation’s potential ability to exert social control, as well as limitations thereof, and hope that this study encourages further work in this vein.

REFERENCES

Abrahamson, E., & Rosenkopf, L. 1993. Institutional and competitive bandwagons: Using mathematical mod- eling as a tool to explore innovation diffusion. Acad- emy of Management Review, 18: 487–517.

Ahmadjian, C. L., & Robinson, P. 2001. Safety in numbers: downsizing and the deinstitutionalization of perma- nent employment in Japan. Administrative Science Quarterly, 46: 622–654.

Arthaud-Day, M., Certo, S. T., Dalton, C. M., & Dalton, D. R. 2006. A changing of the guard: Executive and director turnover following corporate financial restatements. Academy of Management Journal, 49: 1119–1136.

Barnett, M. L. 2013. Why stakeholders ignore firm mis- conduct: A cognitive view. Journal of Management. Published online ahead of print.

Bebchuk, L. A., & Fried, J. M. 2006. Pay without perfor- mance: The unfulfilled promise of executive com- pensation. Cambridge, MA: Harvard University Press.

Bednar, M. K. 2012. Watchdog or lapdog? A behavioral view of the media as a governance mechanism. Academy of Management Journal, 55: 131–150.

Bednar, M. K., Boivie, S., & Prince, N. R. 2013. Burr under the saddle: How media coverage influences strategic change. Organization Science, 24: 910–925.

Berger, P. G., Ofek, E., & Yermack, D. L. 1997. Managerial entrenchment and capital structure decisions. The Journal of Finance, 52: 1411–1438.

Berle, A. A., & Means, G. C. 1932. The modern corporation and private property. New York, NY: Macmillan.

Blair,M. M., & Stout, L. A. 1999. A team productiontheoryof corporate law. Virginia Law Review, 85(2): 247–328.

Boivie, S., Lange, D., McDonald, M. L., & Westphal, J. D. 2011. Me or we: The effects of CEO organizational

2015 1757Bednar, Love, and Kraatz

identification on agency costs. Academy of Man- agement Journal, 54: 551–576.

Bromley, D. B. 1993. Reputation, image and impression management. London, UK: John Wiley & Sons.

Brown, B., & Perry, S. 1994. Removing the financial per- formance halo from fortunes most admired companies. Academy of Management Journal, 37: 1347–1359.

Burt, R.S.2005. Brokerageandclosure:Anintroductionto social capital. Oxford, UK: Oxford University Press.

Burt,R.S.2007.Closure and stability: Persistent reputation and enduring relations among bankers and analysts. The missing links: Formation and decay of economic networks. New York, NY: Russell Sage Foundation.

Chatterjee, A., & Hambrick, D. C. 2007. It’s all about me: Narcissistic chief executive officers and their effects on company strategy and performance. Administra- tive Science Quarterly, 52: 351–386.

Coffee, J. C. 2006. Gatekeepers: The professions and corpo- rate governance. Oxford, UK: Oxford University Press.

Coleman, J. S. 1988. Social capital in the creation of human capital. American Journal of Sociology, 94: 95–120.

Core, J. E., Guay, W., & Larcker, D. F. 2008. The power of the pen and executive compensation. Journal of Finan- cial Economics, 88: 1–25.

Cowen, A. P., & Marcel, J. J. 2011. Damaged goods: Board decisions to dismiss reputationally compromised di- rectors.AcademyofManagement Journal,54:509–527.

Dalton,D.R.,Hitt,M.A.,Certo,S.T.,&Dalton,C.M.2007.The fundamental agency problem and its mitigation: Inde- pendence, equity, and the market for corporate control. The Academy of Management Annals, 1: 1–64.

Davis, G. F. 1991. Agents without principles? The spread of the poison pill through the intercorporate network. Administrative Science Quarterly, 36: 583–613.

Davis, G. F. 2009. Managed by the markets: How finance re-shaped America. New York, NY: Oxford Univer- sity Press.

Davis, G. F., & Greve, H. R. 1997. Corporate elite networks and governance changes in the 1980s. American Journal of Sociology, 103: 1–37.

Dyck, A., & Zingales, L. (Eds.) (2002). The corporate gov- ernance role of the media. Washington, DC: The World Bank Institute.

Ertimur, Y., Ferri, F., & Maber, D. A. 2012. Reputation penalties for poor monitoring of executive pay: Evi- dence from option backdating. Journal of Financial Economics, 104: 118–144.

Fama,E.F.,&Jensen,M.C.1983.Separationofownershipand control. The Journal of Law & Economics, 26: 301–325.

Finkelstein, S., & D’Aveni, R. A. 1994. CEO duality as a double-edged sword: how boards of directors balance

entrenchment avoidance and unity of command. Academy of Management Journal, 37: 1079–1108.

Finkelstein, S., & Hambrick, D. C. 1990. Top-management team tenure and organizational outcomes: the mod- erating role of managerial discretion. Administrative Science Quarterly, 35: 484–503.

Fiske, S. T., & Taylor, S. E. 2013. Social cognition: From brains to culture. Atlanta, GA: Sage.

Fiske, S. T., Cuddy, A. J., & Glick, P. 2007. Universal di- mensions of social cognition: Warmth and compe- tence. Trends in Cognitive Sciences, 11: 77–83.

Fombrun, C. 1996. Reputation: Realizing value from the corporateimage.Boston, MA:HarvardBusinessPress.

Fombrun, C., & Shanley, M. 1990. What’s in a name: Rep- utation building and corporate strategy. Academy of Management Journal, 33: 233.

Freeman, R. E., & Evan, W. M. 1990. Corporate governance: a stakeholder interpretation. The Journal of Behav- ioral Economics, 19: 337–359.

Gilman, H., & Hertzberg, D. 1985, August 20. Pantry pride announces a hostile offer for Revlon, which moves to block bid. Wall Street Journal, 2.

Gomes,A.2000.Goingpublicwithoutgovernance:Managerial reputation effects. The Journal of Finance, 55: 615–646.

Gompers, P., Ishii, J., & Metrick, A. 2003. Corporate gov- ernance and equity prices. The Quarterly Journal of Economics, 118: 107–155.

Graffin, S., Pfarrer, M., & Hill, M. (Eds.) (2012). Untangling executive reputation and corporate reputation: Who made who? Oxford, U.K.: Oxford University Press.

Granovetter, M. 1985. Economic action and social struc- ture: the problem of embeddedness. American Jour- nal of Sociology, 91: 481–510.

Greve, H. R., Palmer, D., & Pozner, J. E. 2010. Organizations gone wild: The causes, processes, and consequences of organizational misconduct. The Academy of Management Annals, 4: 53–107.

Hayward, M. L. A., Rindova, V. P., & Pollock, T. G. 2004. Be- lieving one’s own press: The causes and consequences of CEO celebrity. Strategic Management Journal, 25: 637.

Heckman, J. J. 1979. Sample selection bias as a specifica- tion error. Econometrica, 47: 153–161.

Heider, F. 1958. Psychology of interpersonal relations. New York, NY: Wiley.

Hewstone, M., Rubin, M., & Willis, H. 2002. Intergroup bias. Annual Review of Psychology, 53: 575–604.

Hirsch, P. M. 1986. From ambushes to golden parachutes: corporate takeovers as an instance of cultural framing and institutional integration. American Journal of Sociology, 91: 800–837.

Hsiao, C. 1986. Analysis of panel data. Cambridge, U.K.: Cambridge University Press.

1758 DecemberAcademy of Management Journal

Jaccard, J., Turrisi, R., & Wan, C. K. 1990. Interaction ef- fects in multiple regression. Newbury Park, CA: Sage.

Jensen, M. C. 1988. The takeover controversy: Analysis and evidence. In J. C. Coffee, L. Lowenstein & S. Rose- Ackerman (Eds.), Knights, raiders, and targets: The impact of the hostile takeover: 314–354. New York, NY: Oxford University Press.

Jensen, M. C. 1989. The eclipse of the public corporation. HarvardBusiness Review, September–October:61–74.

Jensen, M. C., & Meckling, W. H. 1976. Theory of the firm: Managerial behavior, agency costs, and own- ership structure. Journal of Financial Economics, 3: 305–350.

Jones,E.E., & Davis, K.E.1965. Fromactsto dispositionsthe attribution process in person perception. Advances in Experimental Social Psychology, 2: 219–266.

Kelley, H. H. 1967. Attribution theory in social psychol- ogy. Paper presented at the Nebraska symposium on motivation.

Kelley, H. H. 1973. The processes of causal attribution. American Psychologist, 28: 107–128.

Kelley, H. H., & Michela, J. L. 1980. Attribution theory and research. Annual Review of Psychology, 31: 457–501.

Kim, H., & Jensen, M. 2014. Audience heterogeneity and the effectiveness of market signals: How to overcome liabilities of foreignness in film exports? Academy of Management Journal. Published online ahead of print.

King, B. G., & Soule, S. A. 2007. Social movements as extra- institutional entrepreneurs: The effect of protests on stock price returns. Administrative Science Quar- terly, 52: 413–442.

Klein, B., & Leffler, K. B. 1981. The role of market forces in assuring contractual performance. Journal of Politi- cal Economy,89: 615–641.

Kraatz, M. S., & Block, E. S. 2008. Organizational Impli- cations of Institutional Pluralism. In R. Greenwood, C. Oliver, R. Suddaby & K. Sahlin-Andresson (Eds.), Handbook of Organizational Institutionalism: 243–275. London, UK: Sage.

Mackie, D. M., & Smith, E. R. 1998. Intergroup relations: insights from a theoretically integrative approach. Psychological Review, 105: 499.

McCoy, C. F., & Williams, J. D. 1985, February 12. Phillips Petroleum’s “poison pill” move is criticized by Pick- ens as entrenchment. Wall Street Journal: 6.

McDonald, M. L., & Westphal, J. D. 2003. Getting by with the advice of their friends: CEOs’ advice networks and firms’ strategic responses to poor performance. Ad- ministrative Science Quarterly, 48: 1–32.

Mitchell, R. K., Agle, B. R., & Wood, D. J. 1997. Toward a theory of stakeholder identification and salience:

Defining the principle of who and what really counts. Academy of Management Review, 22: 853–886.

Morris, M. W., &Peng, K. 1994. Cultureandcause:American and Chinese attributions for social and physical events. Journal of Personality and Social Psychology, 67: 949–971.

Ocasio, W. 1997. Towards an attention-based view of the firm. Strategic Management Journal, 18: 187–206.

Pettigrew, T. F. 1979. The ultimate attribution error: Extend- ing Allport’s cognitive analysis of prejudice. Personality and Social Psychology Bulletin, 5: 461–476.

Pollock, T. G., & Rindova, V. P. 2003. Media legitimation effects in the market for initial public offerings. Academy of Management Journal, 46: 631.

Pollock, T. G., Rindova, V. P., & Maggitti, P. G. 2008. Market watch: Information and availability cascades among the media and investors in the U.S. IPO market. Academy of Management Journal, 51: 335–358.

Rogers, E. M. 1995. Diffusion of innovations. New York, NY: The Free Press.

Ryngaert, M. 1988. The effect of poison pill securities on shareholder wealth. Journal of Financial Economics, 20: 377–417.

Sanders, W. G., & Hambrick, D. C. 2007. Swinging for the fences: The effects of CEO stock options on company risk taking and performance. Academy of Manage- ment Journal, 50: 1055–1078.

Schruijer, S., Blanz, M., Mummendey, A., Tedeschi, J., Banfai, B., Dittmar, H., Kleibaumhüter, P., Mahjoub, A., Mandrosz-Wroblewska, J., Molinari, L., & Petillon, X. 1994. The group-serving bias in evaluating and explaining harmful behavior. The Journal of Social Psychology, 134: 47–53.

Shapiro, C. 1983. Premiums for high quality products as returns to reputations. The Quarterly Journal of Economics, 98: 659–679.

Shaver, J. M. 1998. Accounting for endogeneity when assessing strategy performance: Does entry mode choice affect FDI survival? Management Science, 44: 571–585.

Stewart, J. B., & Hertzberg, D. 1986, January 13. Landmark victory: Outside directors led the carbide defense that fended off GAF—Throughout, they struggled to pro- tect stockholders; Bhopal raised pressure—Harry Gray’s aches and pains. Wall Street Journal.

Tolbert, P. S., & Zucker, L. G. 1983. Institutional sources of change in the formal structure of organizations: The diffusion of civil service reform, 1880-1935. Admin- istrative Science Quarterly, 28: 22–39.

Tsui, A. S. 1984. A role set analysis of managerial reputa- tion. Organizational Behavior and Human Perfor- mance, 34: 64–96.

2015 1759Bednar, Love, and Kraatz

Useem, M. 1984. The inner circle: Large corporations and the rise of business political activity in the US and UK. Oxford, UK: Oxford University Press.

Useem, M. 1993. Executive defense: Shareholder power and corporate reorganization. Cambridge, MA: Harvard University Press.

Uzzi, B. 1999. Embeddedness in the making of financial capital: how social relations and networks benefit firms seeking financing. American Sociological Re- view, 64: 481–505.

Velasco, J. 2001. The enduring illegitimacy of the poison pill. The Journal of Corporation Law, 27: 381.

Wade, J., O’Reilly, C. A., III, & Chandratat, I. 1990. Golden parachutes: CEOs and the exercise of social influence. Administrative Science Quarterly, 35: 587–603.

Wade, J. B., O’Reilly, C. A., III, & Pollock, T. G. 2006. Overpaid CEOs and underpaid managers: Fairness and executive compensation. Organization Science, 17: 527–544.

Walsh, J. P., & Seward, J. K. 1990. On the efficiency of in- ternal and external corporate control mechanisms. Academy of Management Review, 15: 421–458.

Weigelt, K., & Camerer, C. 1988. Reputation and corporate strategy: a review of recent theory and applications. Strategic Management Journal, 9: 443–454.

Wiersema, M. F., & Zhang, Y. 2011. CEO dismissal: The role of investment analysts. Strategic Management Journal, 32: 1161–1182.

Wiesenfeld, B. M., Wurthmann, K. A., & Hambrick, D. C. 2008. The stigmatization and devaluation of elites associated with corporate failures: A process model. Academy of Management Review, 33: 231–251.

Wilder, D. A., & Thompson, J. E. 1980. Intergroup contact with independent manipulations on in-group and out- group interaction. Journal of Personality and Social Psychology, 38: 589.

Wowak, A. J., & Hambrick, D. C. 2010. A model of person- pay interaction: how executives vary in their re- sponses to compensation arrangements. Strategic Management Journal, 31: 803–821.

Zajac, E. J., & Westphal, J. D. 2004. The social construction of market value: institutionalization and learning perspectives on stock market reactions. American Sociological Review, 69(3): 433–457.

Zuckerman, E. W. 1999. The categorical imperative: Secu- rities analysts and the illegitimacy discount. American Journal of Sociology, 104: 1398.

Michael Bednar ([email protected]) is an assistant professor in the Department of Business Administration at the University of Illinois. His research interests include issues related to corporate governance, executive leader- ship, reputation, and the media. He received his PhD from the McCombs School of Business at the University of Texas at Austin.

E. Geoffrey Love ([email protected]) is an assistant pro- fessor in the Department of Business Administration at the University of Illinois. He received his PhD in organiza- tional behavior from Harvard University. His research in- terests focus on organizational reputation and status, the role of categories in institutional processes, organizational downsizing, and organizational change.

Matthew Kraatz ([email protected]) is an associate pro- fessor in the Department of Business Administration at the University of Illinois. His scholarly interests include or- ganizational adaptation, learning, governance, identity, reputation, leadership, and other institutional processes. He received his PhD from the Kellogg Graduate School of Management at Northwestern University.

1760 DecemberAcademy of Management Journal

Copyright of Academy of Management Journal is the property of Academy of Management and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.