# Hostile Resistance to Hedge Fund Activism

Hostile Resistance to Hedge Fund Activism Abstract When facing hedge fund activists, target firms often fight back. Targets with agency problems and those confronting the threat of investor coordination frequently engage in hostile resistance by implementing governance changes associated with managerial entrenchment. The market negatively responds to hostile resistance, and unless hedge funds counterresist, these campaigns have worse operating performance, faster activist exit, and fewer mergers than do campaigns without hostile target resistance. By contrast, when hedge funds counterresist with proxy fights, lawsuits, or unsolicited tender offers, the impact of hostile target resistance is reversed, and these campaigns have similar outcomes to campaigns without hostile target resistance. Hedge fund activism requires attention and warrants similar preparation as to that we recommend for responding to a hostile takeover bid. Martin Lipton, founding partner of Wachtell, Lipton, Rosen & Katz In the past several years, hedge fund activists have targeted thousands of public companies, seeking changes in corporate control, management, capital structure, board composition, and governance. Under siege by an activist, target firm management faces a decision: they can ignore, negotiate with, or resist the activist. When targets resist, they often take hostile action by legally modifying their corporate charters or bylaws to restrict shareholder voting power or by filing lawsuits against hedge funds, actions that can further entrench managers. Responding to this resistance, activists can counterresist by initiating a proxy contest, filing a lawsuit, or making an unsolicited tender offer. In this framework, activism is a sequential process starting with the initial activist overture, which is followed by the target response and finally the activist counterresponse. While a large literature examines hedge fund activism, the prior literature mainly focuses on the role of the activist and not on the target firm’s response. In this paper, we examine how target firm resistance to activism and subsequent hedge fund counterresistance can affect activism outcomes and target firm value. Recent news reports indicate that corporate managers fear activists and spend significant resources to resist them (George and Lorsch 2014). Managers have strong private incentives to resist, including concerns about job security, reputational damage, stagnant employee pay, changes to boards, or a higher probability of a merger.1 These concerns are consistent with empirical findings that more aggressive activism is associated with more dramatic changes at target firms.2 Managers may also believe that activism will harm firm value. Given strong incentives to resist activists and substantial anecdotal but limited empirical evidence of resistance, we investigate target firm resistance and hedge fund responses. More precisely, we address the following questions: In what ways do target firms resist activists? Why do target firms resist activists? Does resistance reduce the potential benefits of activism? If so, can hedge fund counterresistance following target resistance counteract the negative impact of resistance? Our analysis explores the activism process from the target firm’s perspective and also ascertains the impact of target firm resistance and hedge fund counterresistance on activist outcomes and target firm value. We begin by describing and categorizing target firms’ responses to hedge fund activism. In the face of activism, target firms’ reactions convey varying degrees of hostility. For example, filing a lawsuit is a direct and antagonistic response to an activist. By contrast, verbally denouncing the activist is less hostile. Between these two extremes, target firms often resist by worsening governance. Governance changes often include one or more of the six components of the entrenchment index (E index).3 Because past literature finds that these six provisions—as opposed to other governance provisions—have a negative relation with firm value, we classify target resistance that employs these provisions as hostile target resistance.4 We classify less aggressive target actions, such as rejecting or verbally denouncing hedge fund advances, changing advance notice requirements for elections, or postponing or adjourning the annual meeting, as moderate target resistance. We classify the remaining campaigns as having no target resistance. Although we classify these latter campaigns as having no resistance, they almost always involve publicly observable negotiations, such as meetings, phone calls, or letters, between the parties. However, these negotiations do not reach the threshold of either hostile or moderate resistance. Using campaign, hedge fund, and target firm characteristics as predictors, we investigate why target firms resist. Since resistance can be either hostile or moderate, we use multinomial logit models to model both types simultaneously. Target managers are more likely to engage in hostile resistance when the activist wants to buy the target; when activist ownership is higher; and when the activist has engaged in a proxy fight, filed a lawsuit, or made an unsolicited tender offer in a prior campaign. These results also hold for moderate resistance, but the implied resistance probabilities are much lower. Targets are more likely to engage in hostile resistance (but not moderate resistance) when the CEO is the board chair, when the CEO has longer tenure, when insider ownership is lower, and when the target has more cash or more concentrated institutional ownership. Concentrated institutional ownership can facilitate investor coordination since hedge funds need to convince fewer investors to support their campaigns.5 Together, these findings indicate that target firms with high agency problems; entrenched managers; and those facing aggressive activists, the threat of investor coordination, or loss of firm control are more likely to engage in hostile resistance.6 Next, we describe hedge fund counterresistance against hostile target resistance by categorizing activist-initiated proxy fights, unsolicited tender offers, and lawsuits as formal hedge fund counterresistance. In several campaigns, we do not directly observe the hostile target resistance that leads to formal hedge fund counterresistance. Rather, we infer hostile target resistance from the formal hedge fund counterresistance, based on prior literature which assumes that when hedge funds formally counterresist, the earlier private negotiations were hostile and ineffective (see Brav, Jiang, and Kim 2015).7 About two-thirds of campaigns with hostile target resistance involve formal hedge fund counterresistance; in the remainder of hostile resistance campaigns, hedge funds informally respond by threatening a proxy fight, reiterating their demands, or expressing disappointment. We classify these latter campaigns as having no formal hedge fund counterresistance. We find that hedge funds are less likely to formally counterresist when institutional ownership concentration is high. This finding suggests that hedge funds may deem it unnecessary to engage in costly formal counterresistance against firms with concentrated institutional ownership, since it is relatively easy to informally amass support among a smaller number of influential shareholders. Of course, target firms may also find it easier to amass shareholder support when institutional ownership concentration is high. This possibility suggests an alternative explanation for why hedge funds do not formally counterresist: perhaps hedge funds believe that the other large shareholders will side with management. Hedge fund counterresistance is also more likely when the CEO is the board chair, the target firm has a poison pill in place, and insider ownership is low. Figure 1 presents a flowchart of the activism, target resistance, and hedge fund counterresistance decision process and introduces key terminology. Figure 1 View largeDownload slide Stages of resistance and counterresistance Figure 1 View largeDownload slide Stages of resistance and counterresistance We classify target firm resistance to hedge fund activism into three mutually exclusive categories: no target resistance, moderate target resistance, and hostile target resistance. Conditional on hostile target resistance, we classify the hedge fund’s response into two mutually exclusive categories: unopposed hostile resistance, which includes campaigns with no formal hedge fund counterresistance, and opposed hostile resistance, which includes campaigns with formal hedge fund counterresistance. Next, we examine how target firm resistance affects the efficacy of activism. While the market positively reacts to the announcement of activism, the market’s reaction to target firms’ resistance depends on the nature of the resistance. We find no market reaction to moderate resistance. By contrast, the market negatively reacts to hostile resistance and thereby reduces the positive initial market reaction to the announcement of activism. Finally, formal hedge fund counterresistance against hostile target resistance elicits a positive market response. Next, we examine the likelihood of a merger and changes in operating performance for up to 2 years after activism. The probability of a merger is highest for campaigns with opposed hostile resistance, followed by moderate resistance, no resistance, and unopposed hostile resistance. The difference in probability between opposed and unopposed hostile resistance is statistically significant, as is the difference between unopposed hostile resistance and moderate resistance. However, the difference in probability between moderate resistance and no resistance is insignificant. Similarly, changes in operating performance are best for campaigns with opposed hostile resistance and worst for unopposed hostile resistance; the difference between these two subsamples is statistically significant. By contrast, differences in operating performance among opposed hostile resistance, no resistance, and moderate resistance are statistically insignificant. Overall, these results imply that hostile resistance to hedge fund activism and hedge fund counterresistance both appear to have real and lasting impacts on the efficacy of activism. Hostile target resistance is negatively associated with target firms’ operating performance and the probability that a target firm will merge. However, formal counterresistance by hedge funds appears to reduce this negative impact. Since the potential for shareholder coordination—as proxied by institutional ownership concentration—affects a target firm’s decision to resist and a hedge fund’s decision to counterresist, we examine the impact of coordination on target firms. During an activism campaign, a subsequent activist sometimes emerges; in many of these campaigns the activists state explicit support for each other. In campaigns with explicit support, the likelihood of merger is higher and operating performance is better than in campaigns without explicit support. This higher likelihood of a merger provides a potential explanation for our finding that targets resist in a hostile manner when faced with concentrated institutional ownership. Because formal hedge fund counterresistance appears to mitigate the negative impact of hostile target resistance on mergers and operating performance, we investigate why some activists choose not to formally oppose their targets. One possible explanation is that these activists rationally minimize their potential losses by quickly selling shares rather than by engaging in expensive formal counterresistance. Thus, we perform logit regressions in which the dependent variable (called early exit) is set to 1 if the hedge fund exits the firm within 1 year of activism without achieving any of its stated goals and without the firm merging, liquidating, or delisting. Stated activism goals typically include requests for board representation; a firm or asset sale; or governance, management, or operating changes.8 When faced with hostile target resistance, activists that do not formally counterresist are significantly more likely to exit early. However, contrary to the prior literature, early exit by hedge funds is not concentrated among liquid stocks. Rather, while stock liquidity is positively linked to early exit for the full sample (like in the prior literature), hedge funds that do not oppose hostile target firm resistance are less likely to exit liquid stocks early (contrary to the prior literature).9 It thus does not appear that the market’s negative response to hostile resistance is driven by hedge funds exiting early from liquid stocks. The negative reaction is instead consistent with an expectation of overall poor outcomes from activism following hostile resistance. Next, we examine the relation between long-term stock performance and the short-term market response to hedge fund activism, target firm resistance, and hedge fund counterresistance. For all campaigns with observable hedge fund exit (not only those with early exit), we estimate activist holding-period abnormal returns to activism, starting a day before activism announcement and ending a day after activist exit. Our primary measure of performance is buy and hold abnormal returns (BHARs), as recommended by Barber and Lyon (1997). Recognizing that long-run event study returns should be interpreted with caution, we find that all campaigns, except those with unopposed hostile resistance, have positive and statistically significant BHARs regardless of the return measure used. The lower returns for the unopposed hostile resist subsample are consistent with the negative announcement (short-term) cumulative abnormal returns (CARs) to hostile target resistance. By contrast, the average holding-period return for the opposed hostile resist subsample is significantly better, consistent with the positive announcement CAR to formal hedge fund counterresistance.10 Next, we address the issue of endogeneity. An alternative explanation for our finding that unopposed hostile target resistance results in bad target outcomes is that the types of targets that resist would incur bad outcomes in any case. We address this problem in several ways. First, unopposed hostile resistance and opposed hostile resistance both involve hostile resistance; the only difference is activist counterresistance.11 If the alternative explanation is correct, we would expect these campaigns to have similar outcomes, but they do not. Rather, formal hedge fund counterresistance appears to reduce the impact of hostile target resistance. Next, we enumerate the ways in which hostile target resistance directly constrains activists.12 We also argue that the high cost of formal counterresistance should deter hedge funds from doing so unless they expect it to have an impact (Gantchev 2013). In our sample, about 30% of formal counterresistance is definitively successful, in that the hedge fund acquires the firm, wins the proxy fight, or wins the lawsuit. However, like in Boyson, Gantchev, and Shivdasani (2017), our finding that formal counterresistance leads to better operating performance does not depend on the definitive success of counterresistance. Instead, similar to Safieddine and Titman (1999), it appears that formal hedge fund counterresistance may induce managers to make positive changes to decrease the likelihood of additional activism. Finally, our results are robust to propensity score matching on observable campaign and target characteristics. Along with its contribution to the literature on hedge fund activism, our paper adds to several other strands of literature.13 It relates to work in shareholder activism, including Del Guercio and Hawkins (1999) and Gillan and Starks (2000), and to literature on exit, voice, and liquidity by Maug (1998), Admati and Pfleiderer (2009), and Edmans, Fang, and Zur (2013). It also relates to Ryngaert (1988) and Malatesta and Walkling (1988), who report negative stock returns around poison pill announcements. Our paper adds to a burgeoning literature on shareholder coordination (Chakraborty and Gantchev 2013; Appel, Gormley, and Keim 2016; Brav, Dasgupta, and Mathews 2017). Target firms are more likely to resist when institutional ownership is concentrated, implying a higher potential for shareholder coordination. Extending the work of Becht et al. (2017), we show that campaigns in which hedge funds explicitly support each other have better operating performance and a higher probability of merging. Our paper also contributes to a large literature on how the adoption of antitakeover provisions can be explained by managerial entrenchment, including DeAngelo and Rice (1983), Bebchuk, Cohen, and Ferrell (2009), Chemmanur, Paeglis, and Simonyan (2011), Cremers and Ferrell (2014), and Heron and Lie (2015). Finally, our paper adds to a large literature on takeovers. Kini, Kracaw, and Mian (2004) view the takeover market as “an external source of discipline that intercedes when internal control mechanisms are relatively weak or ineffective” (p. 1511). Similarly, activism may be viewed as an external source of discipline that sometimes involves an explicit takeover attempt. Like in the market for corporate control, targets are sometimes cooperative and sometimes confrontational; in general, we find that an activist’s choice to oppose hostile target firm resistance can help discipline target firms’ management. 1. Data 1.1 Activism, resistance, and counterresistance sample Data collection begins with a comprehensive sample of about 2,200 campaigns from 2001 to 2012 from Shark Repellant (SR), a division of FactSet that aggregates activism data. Like in Gantchev (2013), we restrict the sample to “purposeful” activism in which hedge funds pursue explicit goals, such as board seats, changes to capital structure, or mergers, and exclude campaigns in which the activist’s only stated purpose is a belief that the target is undervalued.14 We impose this limit to focus on campaigns most likely to elicit resistance from targets. This restriction reduces the sample to 1,008 campaigns, half the size of the Boyson, Gantchev, and Shivdasani (2017) sample, which does not impose this restriction. We exclude over 300 risk arbitrage campaigns (see Jiang, Li, and Mei Forthcoming) in which activism is initiated after a merger announcement, since we do not wish to confound the analysis of the probability of a target firm merging. Matching these data with CRSP and Compustat brings the sample to 821 campaigns. For each campaign, we classify the target firms’ responses to activism as hostile, moderate, or no resistance. Hostile resistance involves provisions of the Entrenchment index (E index), since these provisions have a negative relation with firm value (Bebchuk, Cohen, and Ferrell 2009). The E index includes 6 of 24 governance provisions from the Gompers, Ishii, and Metrick (2003) (GIM) index: supermajority provisions for charter amendments or mergers, limits to shareholder bylaw amendments, classified boards, poison pills, and golden parachutes. We also classify as hostile target firm lawsuits and bylaw changes limiting shareholder ability to call special meetings or act by written consent. Because hedge funds often rely on other investors’ support, these bylaw changes can directly impede activism. Moderate resistance involves the target firm acknowledging and overtly rejecting hedge fund advances, appointing non-hedge-fund directors when hedge funds are lobbying for board seats, changing advance notice requirements, adjourning meetings, adding derivative disclosure requirements, changing the capital structure in a way that the hedge fund dislikes, or making an acquisition that the hedge fund has previously discouraged. No resistance includes all other campaigns. Although we classify these campaigns as having no resistance because they do not meet the moderate or hostile resistance threshold, they typically include public negotiations between the parties. Hedge funds usually respond to hostile target resistance with some type of counterresistance, which we classify as either formal or informal. Formal counterresistance includes actions subject to legal or regulatory oversight, such as proxy fights, unsolicited tender offers, or lawsuits. All other counterresistance, such as a public expression of disappointment in the target, a proxy threat, or reiteration of activist demands, is informal. We refer to hostile target resistance that is followed by formal hedge fund counterresistance as opposed hostile resistance and to hostile target resistance that is followed by informal hedge fund counterresistance as unopposed hostile resistance. Included in the opposed hostile resistance subsample are several campaigns for which we do not directly observe the target firm’s hostile resistance. Rather, we infer the hostile nature of the target’s resistance from the hedge fund’s formal counterresistance. This approach follows Brav, Jiang, and Kim (2015), who assume that when hedge funds engage in formal counterresistance, the private negotiations predicating the formal counterresistance were unproductive and hostile in nature. The opposed hostile resistance campaigns for which we do not directly observe hostile target resistance do not differ along any observable dimensions from the opposed hostile resistance campaigns for which we directly observe hostile resistance. Figure 2 repeats the flowchart of Figure 1 and also reports the number of campaigns in each subsample. Figure 2 View largeDownload slide Stages of resistance and counterresistance, including the number of campaigns Figure 2 View largeDownload slide Stages of resistance and counterresistance, including the number of campaigns Resistance and counterresistance takes several forms. Adopting or strengthening a poison pill comprises 68% of observable hostile resistance. Lawsuits account for 15%, and limiting shareholder ability to act by written consent or call special meetings accounts for 14%. The remaining provisions of the E index collectively sum to 11%. For moderate resistance, the most common action is verbally rejecting activist advances, composing 60% of moderate resistance events, followed by adding independent directors not approved by the hedge fund at 22%. The remaining types of moderate resistance collectively sum to 19%. The most common type of formal hedge fund counterresistance is a proxy fight, occurring in 80% of the opposed hostile resistance subsample, followed by lawsuits at 37% and unsolicited tender offers at 8%. In all subsamples, the proportions add to more than 100% since targets (hedge funds) can resist (counterresist) with more than one action per campaign. 1.2 Hostile target firm resistance and constraints on hedge funds This paper’s analyses rely on the contention that hostile target resistance can directly impede hedge fund activism. In this section, we present evidence to support this claim, beginning with a brief discussion of poison pills, the most common type of hostile resistance. Poison pills were originally designed to force potential acquirers to negotiate with targets rather than conduct hostile takeovers. A poison pill limits investor ownership to a maximum percentage (the threshold percentage). Ownership above this limit will trigger the poison pill, and all shareholders apart from the triggering shareholder may buy stock at a 50% discount. Bebchuk et al. (2013) argue that poison pills can harm shareholders in two ways: they can prevent shareholders from selling stock at beneficial prices, and they can insulate insiders from engaging with large shareholders, thereby imposing costs on all target shareholders.15 The number of poison pills in U.S. firms has declined from over 1,900 in 2002 to only 420 in 2016, based on data from Shark Repellant. Part of this decline derives from activists encouraging targets to rescind their poison pills.16 However, over this period, the average poison pill threshold also decreased from over 20% to below 15%.17 Consistent with these data, new poison pills adopted as a form of hostile target resistance in our sample have an average threshold of just below 15%. Moreover, 58% of new poison pills have a threshold percentage within 2% of the hedge fund’s reported ownership, thereby directly limiting their ability to purchase additional stock. Further, these activists cannot formally join an U.S. Securities and Exchange Commission (SEC)-defined group with another shareholder owning more than 2% of target stock, since their joint ownership would trigger the poison pill. An SEC-defined group includes two or more investors working together for a common goal that jointly file a 13D form. While pills are most common, other types of hostile resistance directly impede activists. In fifteen cases, the target files a lawsuit against the hedge fund. In four cases, the target firm enacts limits to special meetings after the hedge fund has indicated an intent to call a special meeting. In three cases involving board classification, target firms classify their boards after a hedge fund requests board seats. Together, these results provide evidence that hostile target firm resistance can directly constrain activists. 1.3 Summary statistics Table 1 presents summary statistics for activist campaigns for the full sample and for each subsample. All continuous variables are winsorized at the 1% and 99% levels and are defined in the appendix. We begin by describing campaign characteristics, starting with hedge fund ownership levels. In the full sample, hedge fund activists own 9.6% of target firm shares. The highest level of ownership—13.2%—occurs in the unopposed hostile resist subsample. Despite owning minority stakes, activists wield significant potential influence in target firms by virtue of being either the largest shareholder or the largest activist shareholder in firms they target: untabulated results show that in 42% of campaigns, the hedge fund is the largest shareholder, and in 76% of campaigns the hedge fund is the largest activist shareholder. Even more striking, for cases with either hostile or moderate target resistance (Columns 3 to 5), the hedge fund is the largest owner 52% of the time and the largest activist shareholder 81% of the time. Table 1 Firm and campaign characteristics by resistance category With target resist No target resist Moderate target resist Unopposed hostile resist Opposed hostile resist Differences (1) (2) (3) (4) (5) (2)-(3) (3)-(4) (4)-(5) Number of observations 821 465 149 69 138 NA NA NA Campaign characteristics Percentage owned 9.550 8.967 9.143 13.204 10.125 –0.176 –4.061*** 3.079*** Activist in SEC group dummy 0.073 0.037 0.134 0.159 0.087 –0.097*** –0.025 0.072 Other 13D filer dummy 0.381 0.413 0.376 0.391 0.275 0.037 –0.015 0.116* Other 13D percentage 4.791 5.492 4.657 4.260 2.842 0.835 0.397 1.418 Other 13D filer has similar goal dummy 0.102 0.133 0.094 0.072 0.022 0.039 0.022 0.05 Other explicit support dummy 0.141 0.123 0.168 0.188 0.152 –0.045 –0.02 0.036 Hedge fund offer dummy 0.063 0.024 0.107 0.159 0.101 –0.084*** –0.052 0.058 Other takeover dummy 0.346 0.370 0.309 0.275 0.341 0.061 0.034 –0.066 Hedge fund characteristics Order quartile 2.642 2.667 2.711 2.391 2.609 –0.044 0.320* –0.218 HF prior counterresist dummy 0.357 0.295 0.436 0.348 0.486 –0.141*** 0.088 –0.138* Firm characteristics Cash flow/assets 0.013 0.013 0.010 0.017 0.015 0.003 –0.007 0.002 ROA 0.054 0.056 0.052 0.032 0.063 0.004 0.02 –0.031 Market cap. ($MM) 1021 971 1100 1081 1073 $$-$$129 19 8 Cash 0.216 0.213 0.205 0.286 0.200 0.008 –0.081** 0.086** Tobin’s q 1.855 1.821 1.740 1.852 2.094 0.081 –0.112 –0.242 RND 0.050 0.049 0.058 0.056 0.041 –0.009 0.002 0.015 Dividend yield 0.009 0.008 0.008 0.010 0.013 0 –0.002 –0.003 Leverage 0.275 0.272 0.284 0.226 0.297 –0.012 0.058 –0.071 HHI 0.148 0.153 0.148 0.156 0.126 0.005 –0.008 0.03 Liquidity –0.350 –0.358 –0.284 –0.383 –0.375 –0.074 0.099 –0.008 Institutional ownership 0.572 0.574 0.605 0.511 0.556 –0.031 0.094*** –0.045 Shapley value 0.360 0.351 0.343 0.422 0.383 0.008 –0.079 0.039 CEO tenure (years) 7.364 7.057 8.175 7.029 7.699 –1.118 1.146*** –0.67 CEO is board chair dummy 0.453 0.424 0.429 0.493 0.555 –0.005 –0.064 –0.062 Insider ownership 0.138 0.150 0.143 0.120 0.103 0.007 0.023 0.017 CARs Activism CAR%$$_{{[-1,+1]}}$$ 2.94*** 2.16*** 4.97*** 5.51*** 2.12*** –2.81*** –0.54 3.39** Resistance CAR%$$_{{[-1,+1]}}$$ NA NA 0.70 –2.70*** –2.00** NA 3.40*** –0.70 Counterresistance CAR%$$_{{[-1,+1]}}$$ NA NA NA NA 0.85* NA NA NA Buy and hold return%$$_{\mathrm{[-1\ act, + 1 \ exit]}}$$ 11.66*** 10.57*** 17.99*** –3.87 16.57*** –7.42 21.86** 20.44** Holding pd. CAR%$$_{\mathrm{[-1\ act,+ 1 \ exit]}}$$ 22.17*** 20.51*** 24.81*** 14.22* 29.30*** –4.31 10.59 –15.08 Operating performance 1-year change in CF/assets –0.012* –0.019** 0.008 –0.063** 0.017 –0.027 0.071** –0.081** 2-year change in CF/assets 0.003 –0.001 0.026** –0.045 0.016 –0.027* 0.071** –0.061* 1-year change in ROA –0.008* –0.012* 0.000 –0.026 0.004 –0.012 0.026 –0.030 2-year change in ROA 0.002 –0.003 0.016 –0.024 0.017 –0.019 0.041* –0.041 With target resist No target resist Moderate target resist Unopposed hostile resist Opposed hostile resist Differences (1) (2) (3) (4) (5) (2)-(3) (3)-(4) (4)-(5) Number of observations 821 465 149 69 138 NA NA NA Campaign characteristics Percentage owned 9.550 8.967 9.143 13.204 10.125 –0.176 –4.061*** 3.079*** Activist in SEC group dummy 0.073 0.037 0.134 0.159 0.087 –0.097*** –0.025 0.072 Other 13D filer dummy 0.381 0.413 0.376 0.391 0.275 0.037 –0.015 0.116* Other 13D percentage 4.791 5.492 4.657 4.260 2.842 0.835 0.397 1.418 Other 13D filer has similar goal dummy 0.102 0.133 0.094 0.072 0.022 0.039 0.022 0.05 Other explicit support dummy 0.141 0.123 0.168 0.188 0.152 –0.045 –0.02 0.036 Hedge fund offer dummy 0.063 0.024 0.107 0.159 0.101 –0.084*** –0.052 0.058 Other takeover dummy 0.346 0.370 0.309 0.275 0.341 0.061 0.034 –0.066 Hedge fund characteristics Order quartile 2.642 2.667 2.711 2.391 2.609 –0.044 0.320* –0.218 HF prior counterresist dummy 0.357 0.295 0.436 0.348 0.486 –0.141*** 0.088 –0.138* Firm characteristics Cash flow/assets 0.013 0.013 0.010 0.017 0.015 0.003 –0.007 0.002 ROA 0.054 0.056 0.052 0.032 0.063 0.004 0.02 –0.031 Market cap. ($MM) 1021 971 1100 1081 1073 $$-$$129 19 8 Cash 0.216 0.213 0.205 0.286 0.200 0.008 –0.081** 0.086** Tobin’s q 1.855 1.821 1.740 1.852 2.094 0.081 –0.112 –0.242 RND 0.050 0.049 0.058 0.056 0.041 –0.009 0.002 0.015 Dividend yield 0.009 0.008 0.008 0.010 0.013 0 –0.002 –0.003 Leverage 0.275 0.272 0.284 0.226 0.297 –0.012 0.058 –0.071 HHI 0.148 0.153 0.148 0.156 0.126 0.005 –0.008 0.03 Liquidity –0.350 –0.358 –0.284 –0.383 –0.375 –0.074 0.099 –0.008 Institutional ownership 0.572 0.574 0.605 0.511 0.556 –0.031 0.094*** –0.045 Shapley value 0.360 0.351 0.343 0.422 0.383 0.008 –0.079 0.039 CEO tenure (years) 7.364 7.057 8.175 7.029 7.699 –1.118 1.146*** –0.67 CEO is board chair dummy 0.453 0.424 0.429 0.493 0.555 –0.005 –0.064 –0.062 Insider ownership 0.138 0.150 0.143 0.120 0.103 0.007 0.023 0.017 CARs Activism CAR%$$_{{[-1,+1]}}$$ 2.94*** 2.16*** 4.97*** 5.51*** 2.12*** –2.81*** –0.54 3.39** Resistance CAR%$$_{{[-1,+1]}}$$ NA NA 0.70 –2.70*** –2.00** NA 3.40*** –0.70 Counterresistance CAR%$$_{{[-1,+1]}}$$ NA NA NA NA 0.85* NA NA NA Buy and hold return%$$_{\mathrm{[-1\ act, + 1 \ exit]}}$$ 11.66*** 10.57*** 17.99*** –3.87 16.57*** –7.42 21.86** 20.44** Holding pd. CAR%$$_{\mathrm{[-1\ act,+ 1 \ exit]}}$$ 22.17*** 20.51*** 24.81*** 14.22* 29.30*** –4.31 10.59 –15.08 Operating performance 1-year change in CF/assets –0.012* –0.019** 0.008 –0.063** 0.017 –0.027 0.071** –0.081** 2-year change in CF/assets 0.003 –0.001 0.026** –0.045 0.016 –0.027* 0.071** –0.061* 1-year change in ROA –0.008* –0.012* 0.000 –0.026 0.004 –0.012 0.026 –0.030 2-year change in ROA 0.002 –0.003 0.016 –0.024 0.017 –0.019 0.041* –0.041 This table reports the means for the sample of 821 activism campaigns in Column 1 and for the four subsamples: no target resist, moderate target resist, unopposed hostile resist, and opposed hostile resist. The appendix describes all variables. Firm characteristics are measured the year before activism is announced. All continuous variables are winsorized at the 1% and 99% tails. The last three columns report differences in means between subsamples. These columns also report the statistical significance of each difference based on $$t$$-tests allowing for unequal variance. For CARs, we also test whether these values are statistically different from 0 for each subsample. NA, not applicable; HHI, Herfindahl-Hirschman index. ***, **, and * denotes statistical significance for the 1%, 5%, and 10% levels, respectively. Table 1 Firm and campaign characteristics by resistance category With target resist No target resist Moderate target resist Unopposed hostile resist Opposed hostile resist Differences (1) (2) (3) (4) (5) (2)-(3) (3)-(4) (4)-(5) Number of observations 821 465 149 69 138 NA NA NA Campaign characteristics Percentage owned 9.550 8.967 9.143 13.204 10.125 –0.176 –4.061*** 3.079*** Activist in SEC group dummy 0.073 0.037 0.134 0.159 0.087 –0.097*** –0.025 0.072 Other 13D filer dummy 0.381 0.413 0.376 0.391 0.275 0.037 –0.015 0.116* Other 13D percentage 4.791 5.492 4.657 4.260 2.842 0.835 0.397 1.418 Other 13D filer has similar goal dummy 0.102 0.133 0.094 0.072 0.022 0.039 0.022 0.05 Other explicit support dummy 0.141 0.123 0.168 0.188 0.152 –0.045 –0.02 0.036 Hedge fund offer dummy 0.063 0.024 0.107 0.159 0.101 –0.084*** –0.052 0.058 Other takeover dummy 0.346 0.370 0.309 0.275 0.341 0.061 0.034 –0.066 Hedge fund characteristics Order quartile 2.642 2.667 2.711 2.391 2.609 –0.044 0.320* –0.218 HF prior counterresist dummy 0.357 0.295 0.436 0.348 0.486 –0.141*** 0.088 –0.138* Firm characteristics Cash flow/assets 0.013 0.013 0.010 0.017 0.015 0.003 –0.007 0.002 ROA 0.054 0.056 0.052 0.032 0.063 0.004 0.02 –0.031 Market cap. ($MM) 1021 971 1100 1081 1073 $$-$$129 19 8 Cash 0.216 0.213 0.205 0.286 0.200 0.008 –0.081** 0.086** Tobin’s q 1.855 1.821 1.740 1.852 2.094 0.081 –0.112 –0.242 RND 0.050 0.049 0.058 0.056 0.041 –0.009 0.002 0.015 Dividend yield 0.009 0.008 0.008 0.010 0.013 0 –0.002 –0.003 Leverage 0.275 0.272 0.284 0.226 0.297 –0.012 0.058 –0.071 HHI 0.148 0.153 0.148 0.156 0.126 0.005 –0.008 0.03 Liquidity –0.350 –0.358 –0.284 –0.383 –0.375 –0.074 0.099 –0.008 Institutional ownership 0.572 0.574 0.605 0.511 0.556 –0.031 0.094*** –0.045 Shapley value 0.360 0.351 0.343 0.422 0.383 0.008 –0.079 0.039 CEO tenure (years) 7.364 7.057 8.175 7.029 7.699 –1.118 1.146*** –0.67 CEO is board chair dummy 0.453 0.424 0.429 0.493 0.555 –0.005 –0.064 –0.062 Insider ownership 0.138 0.150 0.143 0.120 0.103 0.007 0.023 0.017 CARs Activism CAR%$$_{{[-1,+1]}}$$ 2.94*** 2.16*** 4.97*** 5.51*** 2.12*** –2.81*** –0.54 3.39** Resistance CAR%$$_{{[-1,+1]}}$$ NA NA 0.70 –2.70*** –2.00** NA 3.40*** –0.70 Counterresistance CAR%$$_{{[-1,+1]}}$$ NA NA NA NA 0.85* NA NA NA Buy and hold return%$$_{\mathrm{[-1\ act, + 1 \ exit]}}$$ 11.66*** 10.57*** 17.99*** –3.87 16.57*** –7.42 21.86** 20.44** Holding pd. CAR%$$_{\mathrm{[-1\ act,+ 1 \ exit]}}$$ 22.17*** 20.51*** 24.81*** 14.22* 29.30*** –4.31 10.59 –15.08 Operating performance 1-year change in CF/assets –0.012* –0.019** 0.008 –0.063** 0.017 –0.027 0.071** –0.081** 2-year change in CF/assets 0.003 –0.001 0.026** –0.045 0.016 –0.027* 0.071** –0.061* 1-year change in ROA –0.008* –0.012* 0.000 –0.026 0.004 –0.012 0.026 –0.030 2-year change in ROA 0.002 –0.003 0.016 –0.024 0.017 –0.019 0.041* –0.041 With target resist No target resist Moderate target resist Unopposed hostile resist Opposed hostile resist Differences (1) (2) (3) (4) (5) (2)-(3) (3)-(4) (4)-(5) Number of observations 821 465 149 69 138 NA NA NA Campaign characteristics Percentage owned 9.550 8.967 9.143 13.204 10.125 –0.176 –4.061*** 3.079*** Activist in SEC group dummy 0.073 0.037 0.134 0.159 0.087 –0.097*** –0.025 0.072 Other 13D filer dummy 0.381 0.413 0.376 0.391 0.275 0.037 –0.015 0.116* Other 13D percentage 4.791 5.492 4.657 4.260 2.842 0.835 0.397 1.418 Other 13D filer has similar goal dummy 0.102 0.133 0.094 0.072 0.022 0.039 0.022 0.05 Other explicit support dummy 0.141 0.123 0.168 0.188 0.152 –0.045 –0.02 0.036 Hedge fund offer dummy 0.063 0.024 0.107 0.159 0.101 –0.084*** –0.052 0.058 Other takeover dummy 0.346 0.370 0.309 0.275 0.341 0.061 0.034 –0.066 Hedge fund characteristics Order quartile 2.642 2.667 2.711 2.391 2.609 –0.044 0.320* –0.218 HF prior counterresist dummy 0.357 0.295 0.436 0.348 0.486 –0.141*** 0.088 –0.138* Firm characteristics Cash flow/assets 0.013 0.013 0.010 0.017 0.015 0.003 –0.007 0.002 ROA 0.054 0.056 0.052 0.032 0.063 0.004 0.02 –0.031 Market cap. ($MM) 1021 971 1100 1081 1073 $$-$$129 19 8 Cash 0.216 0.213 0.205 0.286 0.200 0.008 –0.081** 0.086** Tobin’s q 1.855 1.821 1.740 1.852 2.094 0.081 –0.112 –0.242 RND 0.050 0.049 0.058 0.056 0.041 –0.009 0.002 0.015 Dividend yield 0.009 0.008 0.008 0.010 0.013 0 –0.002 –0.003 Leverage 0.275 0.272 0.284 0.226 0.297 –0.012 0.058 –0.071 HHI 0.148 0.153 0.148 0.156 0.126 0.005 –0.008 0.03 Liquidity –0.350 –0.358 –0.284 –0.383 –0.375 –0.074 0.099 –0.008 Institutional ownership 0.572 0.574 0.605 0.511 0.556 –0.031 0.094*** –0.045 Shapley value 0.360 0.351 0.343 0.422 0.383 0.008 –0.079 0.039 CEO tenure (years) 7.364 7.057 8.175 7.029 7.699 –1.118 1.146*** –0.67 CEO is board chair dummy 0.453 0.424 0.429 0.493 0.555 –0.005 –0.064 –0.062 Insider ownership 0.138 0.150 0.143 0.120 0.103 0.007 0.023 0.017 CARs Activism CAR%$$_{{[-1,+1]}}$$ 2.94*** 2.16*** 4.97*** 5.51*** 2.12*** –2.81*** –0.54 3.39** Resistance CAR%$$_{{[-1,+1]}}$$ NA NA 0.70 –2.70*** –2.00** NA 3.40*** –0.70 Counterresistance CAR%$$_{{[-1,+1]}}$$ NA NA NA NA 0.85* NA NA NA Buy and hold return%$$_{\mathrm{[-1\ act, + 1 \ exit]}}$$ 11.66*** 10.57*** 17.99*** –3.87 16.57*** –7.42 21.86** 20.44** Holding pd. CAR%$$_{\mathrm{[-1\ act,+ 1 \ exit]}}$$ 22.17*** 20.51*** 24.81*** 14.22* 29.30*** –4.31 10.59 –15.08 Operating performance 1-year change in CF/assets –0.012* –0.019** 0.008 –0.063** 0.017 –0.027 0.071** –0.081** 2-year change in CF/assets 0.003 –0.001 0.026** –0.045 0.016 –0.027* 0.071** –0.061* 1-year change in ROA –0.008* –0.012* 0.000 –0.026 0.004 –0.012 0.026 –0.030 2-year change in ROA 0.002 –0.003 0.016 –0.024 0.017 –0.019 0.041* –0.041 This table reports the means for the sample of 821 activism campaigns in Column 1 and for the four subsamples: no target resist, moderate target resist, unopposed hostile resist, and opposed hostile resist. The appendix describes all variables. Firm characteristics are measured the year before activism is announced. All continuous variables are winsorized at the 1% and 99% tails. The last three columns report differences in means between subsamples. These columns also report the statistical significance of each difference based on $$t$$-tests allowing for unequal variance. For CARs, we also test whether these values are statistically different from 0 for each subsample. NA, not applicable; HHI, Herfindahl-Hirschman index. ***, **, and * denotes statistical significance for the 1%, 5%, and 10% levels, respectively. Additional campaign characteristics include variables capturing the potential for hedge funds to coordinate with other activists. Brav, Dasgupta, and Mathews (2017) develop a model of “wolf pack” activism by which large shareholders coordinate. Becht et al. (2017) find that about 25% of activism campaigns involve more than one hedge fund and that these engagements have better performance. We expect that hostile target resistance can limit the efficacy of shareholder coordination by placing limits on shareholding voting and shareholders’ ability to call meetings or act by written consent. To test this idea directly, we create the following variables. Activist in SEC-defined group dummy is set to 1 if the activist is part of an SEC-defined group. We ensure that groups consist of separate hedge fund management firms, and not two or more hedge funds from the same firm. Other 13D filer dummy is set to 1 if another 13D filer is active in the target firm at the same time as the hedge fund. Other 13D percentage is the total percentage owned by other 13D filers active in the target firm at the same time as the hedge fund. Other 13D filer has similar goal dummy is set to 1 if at least one other 13D filer states goals similar to those of the hedge fund of interest, but no filer explicitly mentions another filer. Other 13D filer supports activist dummy is set to 1 if another 13D filer states support for and names the hedge fund or if the hedge fund states support for and names another 13D filer.18 We exclude 13D filers that are other corporations or target insiders; most other 13D filers are activist hedge funds. Activists file as an SEC-defined group in about 7% of campaigns. This proportion is sharply higher for campaigns with moderate or hostile target resistance. About 38% of campaigns have at least one other independent 13D filer. Conditional on having another independent 13D filer (not tabulated), average other filer ownership is 13%, indicating highly concentrated 13D filer ownership for these firms. Other independent 13D filers state similar goals without explicit support about 10% of the time and state explicit support 14% of the time. When conditioning on the presence of another 13D filer, these (untabulated) results are more dramatic: other 13D filers state similar goals without explicit support 27% of the time and state explicit support for one another 36% of the time. These findings suggest that the combined support of activists could pose a substantial threat to target firm management. Comparing subsamples, the unopposed hostile resist subsample has more campaigns with another independent 13D filer involved, at 39% relative to 28% for the opposed hostile resist subsample, indicating that when other filers are involved, hedge funds are less likely to formally counterresist, possibly because they believe that the other filers will support their campaign, making formal counterresistance less necessary. The final two campaign characteristics reported in Table 1 indicate whether the activist campaign is explicitly takeover related. Hedge fund offer dummy is set to 1 if the hedge fund makes or wants to make an offer for the firm and includes but is not limited to unsolicited tender offers by hedge funds. Other takeover dummy is set to 1 if the hedge fund wants the firm to be acquired or wants to explore strategic alternatives but does not offer to buy the firm itself. The incidence of hedge funds making offers or stating that they want to make offers is significantly higher in the subsamples of campaigns with resistance, indicating that target management is more likely to resist in these campaigns. We also consider hedge fund characteristics. The variable Order quartile is based on a count of the hedge fund’s activism events from 1994 through the event of interest. Order quartile ranges from 1 to 4, with 4 indicating more experience. Prior counterresist dummy is set to 1 if the hedge fund manager has engaged in formal counterresistance in prior campaigns. This variable captures a hedge fund’s reputation for past aggressive behavior in activism. Table 1 indicates that targeted managers are more likely to resist hedge funds with a history of such behavior. All firm characteristics are measured in the year before activism and are defined in the appendix. The Shapley value, a variable that estimates ownership concentration, is not commonly used in activism research, so we define it here. If activists rely on support from other institutions, ownership concentration could affect a target firm’s (activist’s) decision to resist (counterresist). We estimate the Shapley value using the generalized pivotal player approach for infinite person games like in Milnor and Shapley (1978) for each institutional shareholder owning a stake of at least 3%. The Shapley value is the probability that in a randomly permuted ordering of all shareholders, the given shareholder and her predecessors together have a majority vote but her predecessors alone do not. We collect institutional ownership data from quarterly 13F filings, and for each campaign we sum the Shapley values of all investors owning 3% or more of the company shares, excluding insiders, to obtain a total Shapley value for the firm. A higher Shapley value may imply that activists will find it easier to garner support for their agendas. For the full sample, cash flows and return on assets (ROA) are positive, average market capitalization is $\$$1 billion, cash/assets is 22%, Tobin’s q is a healthy 1.9, leverage is 28%, and institutional ownership is about 57%. These results are consistent with the finding of Brav et al. (2008) that hedge funds target small to mid-cap firms with strong balance sheets and good operating performance. Cash is higher for the unopposed hostile resist subsample relative to all other subsamples. Table 1 also reports 3-day CARs around activism, resistance, and counterresistance beginning 1 day before and ending 1 day after the announcement and excluding campaigns with confounding events. The appendix describes the methodology used to calculate the CARs. For the full sample, the 3-day activism CAR is about 3%, comparable to that in Boyson, Gantchev, and Shivdasani (2017), who study a similar time period. The market responds negatively to hostile resistance by the target. Three-day resistance CARs average a statistically insignificant 0.7% for the moderate resist subsample and a statistically significant$$-$$2.7% and$$-$$2.0% for the unopposed hostile and opposed hostile resist subsamples, respectively. The 3.4% difference between the unopposed hostile resist and moderate resist campaigns is highly significant. The 3-day CAR for formal counterresistance is a statistically significant 0.9%. The market, however, does not appear to anticipate formal hedge fund counterresistance: the 0.7% difference between resistance CARs for the unopposed and opposed hostile resist campaigns is insignificant. We also report activist holding-period returns to activism, beginning 1 day before activism is announced and ending 1 day after the hedge fund exits the campaign. Using daily data, we calculate BHARs and holding-period CARs, each measured in excess of the value-weighted CRSP index. The BHARs are lower than the holding-period CARs and are arguably more appropriate for a long term analysis, as discussed in Barber and Lyon (1997). Hence, in the remainder of the paper, we focus on the BHARs, reporting the long-term CARs for robustness. The BHARs that we report are consistent with those reported in prior activism literature.19 Examining differences across subsamples, holding-period returns are lowest for campaigns with unopposed hostile resistance, with an average BHAR of$$-$$3.8% (which is statistically indistinguishable from zero). By contrast, campaigns with hostile target resistance followed by formal hedge fund counterresistance (i.e., opposed hostile resist) have a statistically significant average BHAR of 17%. BHARs for the other two subsamples are positive and significant. Table 1 also reports changes in ROA and cash flows/assets for the subsamples. Changes in operating performance are highest for the opposed hostile resist campaigns. The higher long-run returns and better operating performance for the opposed hostile resist subsample are notable when compared to the net initial market reaction (3-day CAR) to activism (2.1%), to target resistance ($$-$$2.0%), and to hedge fund counterresistance (0.9%).20 The sum of these three returns is about 1%. By comparison, for the unopposed hostile resist subsample, the net effect of the 3-day announcement returns to activism (5.5%) and target resistance ($$-$$2.7%) is about 2.8%. These net positive announcement returns for both subsamples imply that initial market expectations for the two groups are similar. However, long-term returns for the opposed hostile resist subsample are sharply higher than for the unopposed hostile resist subsample. These net return results imply that the market does not fully predict the magnitude of either the future negative impact of unopposed hostile target resistance or the future positive impact of formal hedge fund counterresistance. 2. Resistance Decisions 2.1 The target’s decision to resist We begin by modeling the probability that the target will resist the activist. Table 2, panel A, combines moderate and hostile resistance in a logit regression, with the dependent variable set to 1 if the target firm engages in either type of resistance and 0 otherwise, including only campaigns with observable target firm resistance.21 We also exclude campaigns with existing poison pills (the most common type of hostile resistance) because targets in these campaigns cannot resist by implementing a new pill.22 In Column 1, these requirements reduce the sample to 425 campaigns. Column 2 excludes an additional 99 campaigns with missing data for the Shapley value, CEO tenure, or CEO is board chair. In both panels of Table 2, instead of regression coefficients, the tables report the absolute change in the probability of resistance for a move from 0 to 1 for the indicator variables and from the 25th percentile to the 75th percentile for the continuous variables, with all other variables set to their means. Regressions include year and one-digit SIC industry dummies; standard errors are clustered by year and hedge fund; and z-values are reported in parentheses. For reference, the unconditional probability of either type of resistance is 36%. Table 2 Decision to resist activists A. Logit model, dependent variable$$=$$1 if target firm resists and 0 otherwise Any resist (1) (2) Campaign characteristics Hedge fund offer dummy 0.490*** 0.480*** (4.47) (4.84) Other takeover dummy 0.031 0.077 (0.73) (1.15) Percentage owned 0.077*** 0.126*** (3.02) (4.41) Hedge fund in SEC group dummy 0.156* 0.232*** (1.68) (2.92) Hedge fund characteristics Order quartile –0.122 –0.161 (–1.58) (–1.40) HF prior counterresistance dummy 0.199*** 0.237*** (3.08) (2.82) Firm characteristics Cash flow/assets 0.008 (0.30) log of market cap. ( \$$MM) 0.112** (2.16) Cash 0.095*** (2.62) Tobin’s q –0.020 (–1.33) RND –0.012 (–0.32) Dividend yield 0.035*** (2.45) Leverage 0.091** (2.09) HHI –0.027 (–0.98) Liquidity –0.009*** (–2.50) Institutional ownership –0.010 (–0.13) Shapley value 0.023 (0.32) log CEO tenure 0.124*** (2.89) CEO is board chair dummy 0.001 (0.09) Insider ownership –0.052 (–1.41) N 425 326 Includes year & industry dummies? Yes Yes Pseudo R$$^{\mathrm{2}}$$ 0.121 0.209 A. Logit model, dependent variable $$=$$ 1 if target firm resists and 0 otherwise Any resist (1) (2) Campaign characteristics Hedge fund offer dummy 0.490*** 0.480*** (4.47) (4.84) Other takeover dummy 0.031 0.077 (0.73) (1.15) Percentage owned 0.077*** 0.126*** (3.02) (4.41) Hedge fund in SEC group dummy 0.156* 0.232*** (1.68) (2.92) Hedge fund characteristics Order quartile –0.122 –0.161 (–1.58) (–1.40) HF prior counterresistance dummy 0.199*** 0.237*** (3.08) (2.82) Firm characteristics Cash flow/assets 0.008 (0.30) log of market cap. ($\$$MM) 0.112** (2.16) Cash 0.095*** (2.62) Tobin’s q –0.020 (–1.33) RND –0.012 (–0.32) Dividend yield 0.035*** (2.45) Leverage 0.091** (2.09) HHI –0.027 (–0.98) Liquidity –0.009*** (–2.50) Institutional ownership –0.010 (–0.13) Shapley value 0.023 (0.32) log CEO tenure 0.124*** (2.89) CEO is board chair dummy 0.001 (0.09) Insider ownership –0.052 (–1.41) N 425 326 Includes year & industry dummies? Yes Yes Pseudo R$$^{\mathrm{2}}$$0.121 0.209 B. Multinomial logit model, dependent variable$$=$$0 if no resist (base case), 1 if moderate target resist, 2 if hostile target resist Specification 1 Specification 2 Moderate Hostile Moderate Hostile Campaign characteristics Hedge fund offer dummy 0.038** 0.523*** 0.106** 0.313*** (2.38) (4.08) (2.16) (4.41) Other takeover dummy 0.000 0.031 0.020 0.023 (0.16) (0.72) (0.73) (0.87) Percentage owned 0.006 0.063*** 0.027** 0.035*** (1.11) (4.24) (2.33) (3.97) Hedge fund in SEC group dummy 0.054 0.062 0.108** 0.063 (1.34) (1.07) (2.11) (1.56) Hedge fund characteristics Order quartile 0.052 –0.223*** 0.062 –0.163** (0.85) (–3.55) (1.16) (–4.09) HF prior counterresistance dummy 0.036** 0.160*** 0.054** 0.087*** (1.97) (3.23) (1.93) (2.72) Firm characteristics Cash flow/assets –0.006 0.012 (–0.49) (1.40) log of market cap. ( \$$MM) –0.035 0.068*** (–0.89) (2.84) Cash –0.002 0.050*** (0.13) (3.32) Tobin’s q –0.005 –0.009 (–0.52) (–1.35) RND –0.006 0.001 (–0.25) (0.06) Dividend yield 0.005 0.012*** (0.83) (2.63) Leverage 0.017 0.025** (1.20) (1.99) HHI 0.004 –0.014* (–0.49) (–1.65) Liquidity 0.000 –0.008** (–0.14) (–2.02) Institutional ownership 0.144*** –0.117*** (3.14) (–3.45) Shapley value –0.101*** 0.100*** (–2.47) (3.17) log CEO tenure 0.036** 0.040* (1.73) (1.87) CEO is board chair dummy –0.022 0.052** (–0.63) (1.70) Insider ownership –0.020 –0.047** (–1.60) (–1.94) N 425 326 Includes year & industry dummies? Yes Yes Pseudo R$$^{\mathrm{2}}$$ 0.102 0.319 B. Multinomial logit model, dependent variable $$=$$ 0 if no resist (base case), 1 if moderate target resist, 2 if hostile target resist Specification 1 Specification 2 Moderate Hostile Moderate Hostile Campaign characteristics Hedge fund offer dummy 0.038** 0.523*** 0.106** 0.313*** (2.38) (4.08) (2.16) (4.41) Other takeover dummy 0.000 0.031 0.020 0.023 (0.16) (0.72) (0.73) (0.87) Percentage owned 0.006 0.063*** 0.027** 0.035*** (1.11) (4.24) (2.33) (3.97) Hedge fund in SEC group dummy 0.054 0.062 0.108** 0.063 (1.34) (1.07) (2.11) (1.56) Hedge fund characteristics Order quartile 0.052 –0.223*** 0.062 –0.163** (0.85) (–3.55) (1.16) (–4.09) HF prior counterresistance dummy 0.036** 0.160*** 0.054** 0.087*** (1.97) (3.23) (1.93) (2.72) Firm characteristics Cash flow/assets –0.006 0.012 (–0.49) (1.40) log of market cap. ( $\$$MM) –0.035 0.068*** (–0.89) (2.84) Cash –0.002 0.050*** (0.13) (3.32) Tobin’s q –0.005 –0.009 (–0.52) (–1.35) RND –0.006 0.001 (–0.25) (0.06) Dividend yield 0.005 0.012*** (0.83) (2.63) Leverage 0.017 0.025** (1.20) (1.99) HHI 0.004 –0.014* (–0.49) (–1.65) Liquidity 0.000 –0.008** (–0.14) (–2.02) Institutional ownership 0.144*** –0.117*** (3.14) (–3.45) Shapley value –0.101*** 0.100*** (–2.47) (3.17) log CEO tenure 0.036** 0.040* (1.73) (1.87) CEO is board chair dummy –0.022 0.052** (–0.63) (1.70) Insider ownership –0.020 –0.047** (–1.60) (–1.94) N 425 326 Includes year & industry dummies? Yes Yes Pseudo R$$^{\mathrm{2}}$$0.102 0.319 Panel A reports the results of logit analyses modeling the target firm’s decision to resist an activist. The dependent variable (Any resist) is set to 1 if a target firm resists in either a moderate or a hostile manner, and 0 otherwise. The regressions in both panels include all campaigns with directly observable target firm resistance and exclude campaigns in which target firms have a poison pill in force prior to the campaign. Panel B reports the results of multinomial logit regressions in which the dependent variable is set to 0 if the campaign involves no observed resistance (base case), 1 if the campaign involves moderate target resistance, and 2 if the campaign involves hostile target resistance. In both panels, specification 1 includes the campaign and the hedge fund characteristics. Specification 2 includes all control variables. For each specification in panel B, the coefficients corresponding to moderate target resistance are reported in the first column, and the coefficients corresponding to hostile target resistance are reported in the second column. Although the results are reported in two columns, each specification represents just one multinomial logit regression. In both panels, the reported values represent the probability change for a move from 0 to 1 for indicator variables and from the 25th to the 75th percentile for continuous variables, setting all other variables at their means. The appendix defines all control variables. Regressions include year and industry (1-digit SIC) dummies. Standard errors are clustered by time and hedge fund, and z-values are reported below the coefficients in parentheses. HHI, Herfindahl-Hirschman index. ***, **, and * denotes statistical significance for the 1%, 5%, and 10% levels, respectively. Table 2 Decision to resist activists A. Logit model, dependent variable$$=$$1 if target firm resists and 0 otherwise Any resist (1) (2) Campaign characteristics Hedge fund offer dummy 0.490*** 0.480*** (4.47) (4.84) Other takeover dummy 0.031 0.077 (0.73) (1.15) Percentage owned 0.077*** 0.126*** (3.02) (4.41) Hedge fund in SEC group dummy 0.156* 0.232*** (1.68) (2.92) Hedge fund characteristics Order quartile –0.122 –0.161 (–1.58) (–1.40) HF prior counterresistance dummy 0.199*** 0.237*** (3.08) (2.82) Firm characteristics Cash flow/assets 0.008 (0.30) log of market cap. ( \$$MM) 0.112** (2.16) Cash 0.095*** (2.62) Tobin’s q –0.020 (–1.33) RND –0.012 (–0.32) Dividend yield 0.035*** (2.45) Leverage 0.091** (2.09) HHI –0.027 (–0.98) Liquidity –0.009*** (–2.50) Institutional ownership –0.010 (–0.13) Shapley value 0.023 (0.32) log CEO tenure 0.124*** (2.89) CEO is board chair dummy 0.001 (0.09) Insider ownership –0.052 (–1.41) N 425 326 Includes year & industry dummies? Yes Yes Pseudo R$$^{\mathrm{2}}$$ 0.121 0.209 A. Logit model, dependent variable $$=$$ 1 if target firm resists and 0 otherwise Any resist (1) (2) Campaign characteristics Hedge fund offer dummy 0.490*** 0.480*** (4.47) (4.84) Other takeover dummy 0.031 0.077 (0.73) (1.15) Percentage owned 0.077*** 0.126*** (3.02) (4.41) Hedge fund in SEC group dummy 0.156* 0.232*** (1.68) (2.92) Hedge fund characteristics Order quartile –0.122 –0.161 (–1.58) (–1.40) HF prior counterresistance dummy 0.199*** 0.237*** (3.08) (2.82) Firm characteristics Cash flow/assets 0.008 (0.30) log of market cap. ($\$$MM) 0.112** (2.16) Cash 0.095*** (2.62) Tobin’s q –0.020 (–1.33) RND –0.012 (–0.32) Dividend yield 0.035*** (2.45) Leverage 0.091** (2.09) HHI –0.027 (–0.98) Liquidity –0.009*** (–2.50) Institutional ownership –0.010 (–0.13) Shapley value 0.023 (0.32) log CEO tenure 0.124*** (2.89) CEO is board chair dummy 0.001 (0.09) Insider ownership –0.052 (–1.41) N 425 326 Includes year & industry dummies? Yes Yes Pseudo R$$^{\mathrm{2}}$$0.121 0.209 B. Multinomial logit model, dependent variable$$=$$0 if no resist (base case), 1 if moderate target resist, 2 if hostile target resist Specification 1 Specification 2 Moderate Hostile Moderate Hostile Campaign characteristics Hedge fund offer dummy 0.038** 0.523*** 0.106** 0.313*** (2.38) (4.08) (2.16) (4.41) Other takeover dummy 0.000 0.031 0.020 0.023 (0.16) (0.72) (0.73) (0.87) Percentage owned 0.006 0.063*** 0.027** 0.035*** (1.11) (4.24) (2.33) (3.97) Hedge fund in SEC group dummy 0.054 0.062 0.108** 0.063 (1.34) (1.07) (2.11) (1.56) Hedge fund characteristics Order quartile 0.052 –0.223*** 0.062 –0.163** (0.85) (–3.55) (1.16) (–4.09) HF prior counterresistance dummy 0.036** 0.160*** 0.054** 0.087*** (1.97) (3.23) (1.93) (2.72) Firm characteristics Cash flow/assets –0.006 0.012 (–0.49) (1.40) log of market cap. ( \$$MM) –0.035 0.068*** (–0.89) (2.84) Cash –0.002 0.050*** (0.13) (3.32) Tobin’s q –0.005 –0.009 (–0.52) (–1.35) RND –0.006 0.001 (–0.25) (0.06) Dividend yield 0.005 0.012*** (0.83) (2.63) Leverage 0.017 0.025** (1.20) (1.99) HHI 0.004 –0.014* (–0.49) (–1.65) Liquidity 0.000 –0.008** (–0.14) (–2.02) Institutional ownership 0.144*** –0.117*** (3.14) (–3.45) Shapley value –0.101*** 0.100*** (–2.47) (3.17) log CEO tenure 0.036** 0.040* (1.73) (1.87) CEO is board chair dummy –0.022 0.052** (–0.63) (1.70) Insider ownership –0.020 –0.047** (–1.60) (–1.94) N 425 326 Includes year & industry dummies? Yes Yes Pseudo R$$^{\mathrm{2}}$$ 0.102 0.319 B. Multinomial logit model, dependent variable $$=$$ 0 if no resist (base case), 1 if moderate target resist, 2 if hostile target resist Specification 1 Specification 2 Moderate Hostile Moderate Hostile Campaign characteristics Hedge fund offer dummy 0.038** 0.523*** 0.106** 0.313*** (2.38) (4.08) (2.16) (4.41) Other takeover dummy 0.000 0.031 0.020 0.023 (0.16) (0.72) (0.73) (0.87) Percentage owned 0.006 0.063*** 0.027** 0.035*** (1.11) (4.24) (2.33) (3.97) Hedge fund in SEC group dummy 0.054 0.062 0.108** 0.063 (1.34) (1.07) (2.11) (1.56) Hedge fund characteristics Order quartile 0.052 –0.223*** 0.062 –0.163** (0.85) (–3.55) (1.16) (–4.09) HF prior counterresistance dummy 0.036** 0.160*** 0.054** 0.087*** (1.97) (3.23) (1.93) (2.72) Firm characteristics Cash flow/assets –0.006 0.012 (–0.49) (1.40) log of market cap. ( $\$$MM) –0.035 0.068*** (–0.89) (2.84) Cash –0.002 0.050*** (0.13) (3.32) Tobin’s q –0.005 –0.009 (–0.52) (–1.35) RND –0.006 0.001 (–0.25) (0.06) Dividend yield 0.005 0.012*** (0.83) (2.63) Leverage 0.017 0.025** (1.20) (1.99) HHI 0.004 –0.014* (–0.49) (–1.65) Liquidity 0.000 –0.008** (–0.14) (–2.02) Institutional ownership 0.144*** –0.117*** (3.14) (–3.45) Shapley value –0.101*** 0.100*** (–2.47) (3.17) log CEO tenure 0.036** 0.040* (1.73) (1.87) CEO is board chair dummy –0.022 0.052** (–0.63) (1.70) Insider ownership –0.020 –0.047** (–1.60) (–1.94) N 425 326 Includes year & industry dummies? Yes Yes Pseudo R$$^{\mathrm{2}}$$0.102 0.319 Panel A reports the results of logit analyses modeling the target firm’s decision to resist an activist. The dependent variable (Any resist) is set to 1 if a target firm resists in either a moderate or a hostile manner, and 0 otherwise. The regressions in both panels include all campaigns with directly observable target firm resistance and exclude campaigns in which target firms have a poison pill in force prior to the campaign. Panel B reports the results of multinomial logit regressions in which the dependent variable is set to 0 if the campaign involves no observed resistance (base case), 1 if the campaign involves moderate target resistance, and 2 if the campaign involves hostile target resistance. In both panels, specification 1 includes the campaign and the hedge fund characteristics. Specification 2 includes all control variables. For each specification in panel B, the coefficients corresponding to moderate target resistance are reported in the first column, and the coefficients corresponding to hostile target resistance are reported in the second column. Although the results are reported in two columns, each specification represents just one multinomial logit regression. In both panels, the reported values represent the probability change for a move from 0 to 1 for indicator variables and from the 25th to the 75th percentile for continuous variables, setting all other variables at their means. The appendix defines all control variables. Regressions include year and industry (1-digit SIC) dummies. Standard errors are clustered by time and hedge fund, and z-values are reported below the coefficients in parentheses. HHI, Herfindahl-Hirschman index. ***, **, and * denotes statistical significance for the 1%, 5%, and 10% levels, respectively. In Column 2, the probability of resistance increases by 48.0% (4,800 basis points) when the hedge fund makes or plans to make an offer. Relative to the unconditional probability of resisting, this result represents a 133% (0.480/0.36) increase. Boyson, Gantchev, and Shivdasani (2017) find that when a hedge fund makes an offer, the probability of acquisition doubles relative to the probability in other activism campaigns, but that hedge fund offers also result in lower acquisition premiums. Hence, this result implies that target firm managers with career concerns, as well as managers concerned about a low offer price, are more likely to resist. By contrast, the mere suggestion that the firm sell itself, represented by the other takeover-related dummy, does not increase the probability of resistance. A move from the 25th percentile to 75th percentile in percentage ownership results in a 35% higher relative likelihood of resistance (0.126/0.36). Resistance is 64% (0.232/0.36) more likely when hedge funds file as an SEC-defined group. Targets are 66% (0.237/0.36) more likely to resist if the hedge fund has a history of formal counterresistance, indicating that hedge fund reputation is an important consideration for target firm management. A move from the 25th percentile to 75th percentile in CEO tenure results in a 34% (0.124/0.36) higher probability of resistance, whereas a move from the 25th percentile to 75th percentile in insider ownership results in a 14% ($$-$$0.052/0.36) lower probability of resistance, although the latter result just misses being statistically significant. These results imply that target firms resist activists when they feel most threatened, as indicated by positive coefficients on hedge fund offers, percentage ownership, SEC group membership, prior counterresistance, and the negative coefficient on insider ownership. Targets with more financial resources (larger firms and firms with more cash, higher dividends, and higher leverage ratios) are also more likely to resist. Since target firms can choose whether to resist in a moderate or a hostile manner, Table 2, panel B, presents multinomial logit regressions modeling this decision process. The dependent variable is set to 0 for no resistance, 1 for moderate resistance, and 2 for hostile resistance. The regressions use the same sample and independent variables used in panel A. The results of each regression are reported in two columns: the first column reports coefficients for moderate resistance and the second column reports coefficient for hostile resistance. The unconditional probability of hostile resistance for campaigns in these regressions is 17%, and the probability of moderate resistance is 19%. Specification 2 shows that hedge fund offer and percentage owned are positively associated with both moderate and hostile resistance. The relative impact of a hedge fund offer on hostile resistance of 184% (0.313/0.17) is three times the relative impact of a hedge fund offer on moderate resistance of 56% (0.106/0.19). Similarly, the relative impact of percentage ownership is higher for hostile resistance compared with moderate resistance; a move from the 25th percentile to 75th percentile in percentage ownership results in a 21% (0.035/0.17) higher likelihood of hostile resistance and a 14% (0.027/0.19) higher likelihood of moderate resistance. Further, the negative relation in panel A between insider ownership and target resistance and the positive relations in panel A between target resistance and target size, target cash, target dividend yield, and target leverage ratio are completely driven by the hostile target resistance subsample. Finally, high CEO tenure is positively associated with both moderate and hostile resistance, and the CEO being the board chair is positively associated with hostile resistance, but not with moderate resistance. Panel B also indicates that managers of target firms consider both the level and concentration of institutional ownership when deciding whether and how to resist activists. Targets are less likely to engage in hostile resistance when institutional ownership levels are high, but more likely when institutional ownership is concentrated. These results imply that managers concerned about potential shareholder coordination react with hostile resistance that directly inhibits shareholder voting power. However, these managers worry less about diffusely held institutional ownership. Our finding contradicts the alternative possibility that higher Shapley values might discourage hostile resistance because management believes that they are likely to be outvoted. By contrast, targeted managers are more likely to engage in moderate resistance when institutional ownership levels are high, but less likely when ownership concentration is high. These results imply that managers choose hostile resistance in favor of moderate resistance when they feel most threatened by the potential for investor coordination. Taken together, these findings provide strong support for the idea that entrenched managers facing direct threats to their careers are more likely to engage in hostile resistance that inhibits shareholder voting power and reduces the ability for shareholders to enact changes through shareholder coordination. 2.2 The hedge fund decision to formally counterresist Table 3 presents a logit model of the hedge fund decision to formally counterresist. Regressions in Table 3 include the 207 campaigns with hostile target resistance, with the dependent variable set to 1 if the hedge fund formally counterresists and 0 otherwise. Column 2 excludes 45 campaigns with missing data for the Shapley value, CEO tenure, or CEO is board chair dummy. Instead of regression coefficients, the table reports the absolute probability change for a move from 0 to 1 for indicator variables and from the 25th percentile to the 75th percentile for continuous variables, with all other variables set to their means. Regressions include year and one-digit SIC industry dummies; standard errors are clustered by year and hedge fund; and z-values are reported in parentheses. The specification is the same as that used in Table 2, except we exclude the Hedge fund offer dummy variable because of a mechanical correlation between some realizations of this variable and the dependent variable.23 As a benchmark, the unconditional probability of formal hedge fund counterresistance is 67% for campaigns in the Column 1 regression, and 70% for campaigns in Column 2. Table 3 Decision for hedge funds to formally counterresist hostile target firm resistance Formal hedge fund counterresistance (1) (2) Campaign characteristics Other takeover dummy 0.040 0.049 (0.49) (0.47) Percentage owned –0.091** –0.115 (–2.20) (–0.59) Hedge fund in SEC group dummy –0.119 –0.057 (–0.85) (–0.87) Hedge fund characteristics Order quartile 0.065 –0.006 (1.49) (–0.06) HF prior counterresistance dummy 0.079 0.067 (1.14) (1.09) Firm characteristics Cash flow/assets –0.041 (–0.99) log of market cap. ( \$$MM) –0.075 (–0.84) Cash –0.080 (–1.18) Tobin’s q –0.025 (–1.08) RND 0.033 (0.85) Dividend yield –0.008 (–0.63) Leverage –0.078 (–0.82) HHI –0.019 (–0.75) Liquidity –0.005 (–0.34) Institutional ownership 0.207*** (2.86) Shapley value –0.194*** (–2.66) log CEO tenure –0.044 (–0.55) CEO is board chair dummy 0.212* (1.89) Insider ownership –0.060* (–1.65) Pill in force 0.178* (1.84) N 207 162 Includes year & industry dummies? Yes Yes Pseudo R$$^{\mathrm{2}}$$ 0.137 0.225 Formal hedge fund counterresistance (1) (2) Campaign characteristics Other takeover dummy 0.040 0.049 (0.49) (0.47) Percentage owned –0.091** –0.115 (–2.20) (–0.59) Hedge fund in SEC group dummy –0.119 –0.057 (–0.85) (–0.87) Hedge fund characteristics Order quartile 0.065 –0.006 (1.49) (–0.06) HF prior counterresistance dummy 0.079 0.067 (1.14) (1.09) Firm characteristics Cash flow/assets –0.041 (–0.99) log of market cap. ($\$$MM) –0.075 (–0.84) Cash –0.080 (–1.18) Tobin’s q –0.025 (–1.08) RND 0.033 (0.85) Dividend yield –0.008 (–0.63) Leverage –0.078 (–0.82) HHI –0.019 (–0.75) Liquidity –0.005 (–0.34) Institutional ownership 0.207*** (2.86) Shapley value –0.194*** (–2.66) log CEO tenure –0.044 (–0.55) CEO is board chair dummy 0.212* (1.89) Insider ownership –0.060* (–1.65) Pill in force 0.178* (1.84) N 207 162 Includes year & industry dummies? Yes Yes Pseudo R$$^{\mathrm{2}}$$0.137 0.225 This table reports the results of logit analyses modeling the hedge fund’s decision to counterresist hostile target firm resistance. The dependent variable is set to 1 if a hedge fund formally resists hostile target firm resistance. The regressions include only campaigns that involve hostile target firm resistance, where the dependent variable equals 1 if a hedge fund formally counterresists and 0 otherwise. The reported values represent the probability change for a move from 0 to 1 for indicator variables and from the 25th to the 75th percentile for continuous variables, setting all other variables at their means. Pill in force is a dummy variable set to 1 if the target firm has a poison pill in place prior to the initiation of hedge fund activism. The appendix describes all other control variables. Regressions include year and industry (1-digit SIC) dummies. Standard errors are clustered by time and hedge fund, and z-values are reported below the coefficients in parentheses. HHI, Herfindahl-Hirschman index. ***, **, and * denotes statistical significance for the 1%, 5%, and 10% levels, respectively. Table 3 Decision for hedge funds to formally counterresist hostile target firm resistance Formal hedge fund counterresistance (1) (2) Campaign characteristics Other takeover dummy 0.040 0.049 (0.49) (0.47) Percentage owned –0.091** –0.115 (–2.20) (–0.59) Hedge fund in SEC group dummy –0.119 –0.057 (–0.85) (–0.87) Hedge fund characteristics Order quartile 0.065 –0.006 (1.49) (–0.06) HF prior counterresistance dummy 0.079 0.067 (1.14) (1.09) Firm characteristics Cash flow/assets –0.041 (–0.99) log of market cap. ( \$$MM) –0.075 (–0.84) Cash –0.080 (–1.18) Tobin’s q –0.025 (–1.08) RND 0.033 (0.85) Dividend yield –0.008 (–0.63) Leverage –0.078 (–0.82) HHI –0.019 (–0.75) Liquidity –0.005 (–0.34) Institutional ownership 0.207*** (2.86) Shapley value –0.194*** (–2.66) log CEO tenure –0.044 (–0.55) CEO is board chair dummy 0.212* (1.89) Insider ownership –0.060* (–1.65) Pill in force 0.178* (1.84) N 207 162 Includes year & industry dummies? Yes Yes Pseudo R$$^{\mathrm{2}}$$ 0.137 0.225 Formal hedge fund counterresistance (1) (2) Campaign characteristics Other takeover dummy 0.040 0.049 (0.49) (0.47) Percentage owned –0.091** –0.115 (–2.20) (–0.59) Hedge fund in SEC group dummy –0.119 –0.057 (–0.85) (–0.87) Hedge fund characteristics Order quartile 0.065 –0.006 (1.49) (–0.06) HF prior counterresistance dummy 0.079 0.067 (1.14) (1.09) Firm characteristics Cash flow/assets –0.041 (–0.99) log of market cap. ( \$$MM) –0.075 (–0.84) Cash –0.080 (–1.18) Tobin’s q –0.025 (–1.08) RND 0.033 (0.85) Dividend yield –0.008 (–0.63) Leverage –0.078 (–0.82) HHI –0.019 (–0.75) Liquidity –0.005 (–0.34) Institutional ownership 0.207*** (2.86) Shapley value –0.194*** (–2.66) log CEO tenure –0.044 (–0.55) CEO is board chair dummy 0.212* (1.89) Insider ownership –0.060* (–1.65) Pill in force 0.178* (1.84) N 207 162 Includes year & industry dummies? Yes Yes Pseudo R$$^{\mathrm{2}}$$0.137 0.225 This table reports the results of logit analyses modeling the hedge fund’s decision to counterresist hostile target firm resistance. The dependent variable is set to 1 if a hedge fund formally resists hostile target firm resistance. The regressions include only campaigns that involve hostile target firm resistance, where the dependent variable equals 1 if a hedge fund formally counterresists and 0 otherwise. The reported values represent the probability change for a move from 0 to 1 for indicator variables and from the 25th to the 75th percentile for continuous variables, setting all other variables at their means. Pill in force is a dummy variable set to 1 if the target firm has a poison pill in place prior to the initiation of hedge fund activism. The appendix describes all other control variables. Regressions include year and industry (1-digit SIC) dummies. Standard errors are clustered by time and hedge fund, and z-values are reported below the coefficients in parentheses. HHI, Herfindahl-Hirschman index. ***, **, and * denotes statistical significance for the 1%, 5%, and 10% levels, respectively. Focusing on Column 2, hedge funds are more likely to formally counterresist when institutional ownership levels are high, but less likely when institutional ownership concentration is high. When institutional ownership levels are high but ownership is diffuse, a proxy contest may be an effective way to communicate with other shareholders despite its higher expense of about \$$ 11 million (Gantchev 2013). However, if ownership is concentrated, then direct communication with other institutions is easier, likely making formal counterresistance less important. Hedge funds are also less likely to formally counterresist when insider ownership is high, potentially because high insider ownership reduces the likelihood of success in formal counterresistance. Taken together, these results imply that hedge funds weigh the costs and benefits when choosing to formally counterresist and select formal counterresistance when the mechanism to communicate with all shareholders is most valuable. 3. Target Firm Outcomes Next, we examine the relation between target firm outcomes, target resistance, and hedge fund counterresistance. Outcomes include stock and operating performance, a merger, and the hedge fund’s exit from the campaign.24 To test for differences in outcomes by resistance subsamples, we create resistance categorical variables, as follows. Any resist is set to 1 for the 356 campaigns that involve moderate or hostile target resistance. Hostile resist is set to 1 for a subset of these: the 207 campaigns that involve hostile target resistance. Opposed hostile resist is set to 1 for a further subset: the 138 campaigns that involve formal hedge fund counterresistance against hostile target resistance. The missing dummy variable is no target resist (465 campaigns). In regressions including just the Any resist dummy, its coefficient estimates the impact of moderate or hostile resistance, relative to no resistance. The interpretation of the coefficients changes when we include all resistance categorical variables, as shown in Equation (1): \begin{align} \textit{Outcome}_{i} & =\alpha_{i}+b_{1}\textit{Any resist}_{i}+b_{2}\textit{Hostile resist}_{i}\notag\\ &\quad +{b}_{3}\textit{Opposed hostile resist}_{i}+\delta X_{i}^{'}+e_{i}. \end{align} (1) In Equation (1), the dependent variable is the outcome for campaign $$i$$; the resistance categorical dummy variables are defined as above; and $$X_{i}^{'}$$ is a vector of industry dummies, time dummies, and hedge fund, target firm, and campaign characteristics. In Equation (1), $$\textit{Any resist}_{i}$$ measures the impact of moderate resistance relative to campaigns with no resistance. $$\textit{Hostile resist}_{i}$$ measures the impact of unopposed hostile resistance relative to campaigns with moderate resistance. Finally, $$\textit{Opposed hostile resist}_{i}$$ measures the impact of opposed hostile resistance relative to campaigns with unopposed hostile resistance. Summing the coefficients in different ways allows comparisons relative to the subset with no resistance. The following chart presents the coefficients and their interpretations. Regression coefficient Interpretation b$$_{\mathrm{1}}$$ Effect of moderate resistance, relative to no resistance b$$_{\mathrm{2}}$$ Incremental effect of hostility b$$_{\mathrm{3}}$$ Incremental effect of formal counterresistance b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}$$ Effect of unopposed hostile resistance, relative to no resistance b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}+$$ b$$_{\mathrm{3}}$$ Effect of opposed hostile resistance, relative to no resistance Regression coefficient Interpretation b$$_{\mathrm{1}}$$ Effect of moderate resistance, relative to no resistance b$$_{\mathrm{2}}$$ Incremental effect of hostility b$$_{\mathrm{3}}$$ Incremental effect of formal counterresistance b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}$$ Effect of unopposed hostile resistance, relative to no resistance b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}+$$ b$$_{\mathrm{3}}$$ Effect of opposed hostile resistance, relative to no resistance Regression coefficient Interpretation b$$_{\mathrm{1}}$$ Effect of moderate resistance, relative to no resistance b$$_{\mathrm{2}}$$ Incremental effect of hostility b$$_{\mathrm{3}}$$ Incremental effect of formal counterresistance b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}$$ Effect of unopposed hostile resistance, relative to no resistance b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}+$$ b$$_{\mathrm{3}}$$ Effect of opposed hostile resistance, relative to no resistance Regression coefficient Interpretation b$$_{\mathrm{1}}$$ Effect of moderate resistance, relative to no resistance b$$_{\mathrm{2}}$$ Incremental effect of hostility b$$_{\mathrm{3}}$$ Incremental effect of formal counterresistance b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}$$ Effect of unopposed hostile resistance, relative to no resistance b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}+$$ b$$_{\mathrm{3}}$$ Effect of opposed hostile resistance, relative to no resistance 3.1 Short-term activism announcement returns Columns 1 and 2 of Table 4 analyze the short-term market reaction (3-day CAR) to activism using ordinary least squares (OLS) regressions. The univariate results in Table 1 show that short-term CARs average about 3% for the entire sample and are highest (at 5.5%) for the unopposed hostile resist subsample. Regressions include all 821 campaigns, apart from 32 that are missing returns data. All regressions in Table 4 include year and one-digit SIC industry dummies; standard errors are clustered by year and hedge fund; and t-values are reported in parentheses. Results from Column 1 indicate that activism announcement CARs for campaigns with any target resistance do not differ from CARs for campaigns with no target resistance. Consistent with the prior literature, CARs are higher when the hedge fund makes an offer to buy the firm or when the hedge fund encourages the target firm to merge. CARs are also higher when the hedge fund files as part of an SEC-defined group. Finally, CARs are higher for firms with better past operating performance and lower for firms with higher Tobin’s q.25 Table 4 Short-term market reaction to activism and resistance Activism CAR$$_{\mathrm{[-1,+1]}}$$ Resistance CAR$$_{\mathrm{[-1,+1]}}$$ (1) (2) (3) (4) Resistance categorical variables Any resist (b$$_{\mathrm{1}})$$ 0.007 0.016* –0.013*** –0.001 (1.37) (1.65) (–2.98) (–0.14) Hostile resist (b$$_{\mathrm{2}})$$ –0.001 –0.038*** (–0.07) (–3.10) Opposed hostile resist (b$$_{\mathrm{3}})$$ –0.024 0.003 (–1.60) (0.28) Campaign characteristics Hedge fund offer dummy 0.076*** 0.075*** 0.020* 0.024** (4.64) (4.39) (1.76) (2.13) Other takeover-related dummy 0.019** 0.019** –0.005 –0.005 (1.97) (2.08) (–0.74) (–0.71) Percentage owned 0.000 0.000 –0.001* 0.000 (0.16) (0.24) (–1.74) (–1.11) Hedge fund in SEC group dummy 0.036** 0.034** 0.019*** 0.021*** (2.35) (2.03) (3.15) (3.19) Other 13D filer dummy –0.005 –0.006 –0.003 –0.003 (–0.50) (–0.61) (–0.78) (–0.85) Other explicit support dummy 0.020 0.021 –0.002 –0.002 (0.98) (1.02) (–0.23) (–0.28) Hedge fund characteristics Order quartile –0.001 –0.001 –0.002 –0.003 (–0.23) (–0.37) (–1.23) (–1.53) HF prior counterresistance dummy 0.005 0.006 0.005 0.004 (0.90) (1.02) (1.02) (0.98) Firm characteristics Cash flow/assets 0.040* 0.040* –0.024 –0.024 (1.84) (1.74) (–0.93) (–0.94) log of market cap. (\$$MM) –0.002 –0.002 0.004 0.004* (–0.71) (–0.72) (1.53) (1.75) Cash 0.007 0.006 0.018 0.022* (0.34) (0.27) (1.43) (1.90) Tobin’s q –0.005*** –0.005*** 0.000 0.000 (–3.80) (–3.88) (–0.07) (–0.07) RND 0.088 0.081 –0.021 –0.032 (1.25) (1.20) (–0.34) (–0.54) Dividend yield –0.023 –0.011 0.015 0.042 (–0.11) (–0.05) (0.10) (0.31) Leverage ratio 0.015 0.016 0.018*** 0.017*** (1.26) (1.32) (7.75) (7.66) HHI –0.005 –0.009 –0.010 –0.010 (–0.18) (–0.33) (–0.45) (–0.48) Liquidity 0.011 0.010 0.000 –0.001 (1.47) (1.47) (–0.25) (–0.61) Institutional ownership 0.003 0.003 –0.008 –0.013 (0.22) (0.23) (–0.87) (–1.33) Insider ownership 0.010 0.007 0.006 –0.002 (0.50) (0.32) (0.68) (–0.21) Pill in force dummy –0.009 –0.009 0.007 –0.024 (–1.50) (–1.42) (1.11) (–0.94) b$$_{\mathrm{1}}+$$b$$_{\mathrm{2}}$$0.015 –0.039*** (1.24) (–3.75) b$$_{\mathrm{1}}+$$b$$_{\mathrm{2}}+$$b$$_{\mathrm{3}}$$–0.008 –0.036*** (–0.93) (–2.49) N 789 789 599 599 Includes year & industry dummies? Yes Yes Yes Yes Adjusted R$$^{\mathrm{2}}$$0.085 0.091 0.008 0.023 Activism CAR$$_{\mathrm{[-1,+1]}}$$Resistance CAR$$_{\mathrm{[-1,+1]}}$$(1) (2) (3) (4) Resistance categorical variables Any resist (b$$_{\mathrm{1}})$$0.007 0.016* –0.013*** –0.001 (1.37) (1.65) (–2.98) (–0.14) Hostile resist (b$$_{\mathrm{2}})$$–0.001 –0.038*** (–0.07) (–3.10) Opposed hostile resist (b$$_{\mathrm{3}})$$–0.024 0.003 (–1.60) (0.28) Campaign characteristics Hedge fund offer dummy 0.076*** 0.075*** 0.020* 0.024** (4.64) (4.39) (1.76) (2.13) Other takeover-related dummy 0.019** 0.019** –0.005 –0.005 (1.97) (2.08) (–0.74) (–0.71) Percentage owned 0.000 0.000 –0.001* 0.000 (0.16) (0.24) (–1.74) (–1.11) Hedge fund in SEC group dummy 0.036** 0.034** 0.019*** 0.021*** (2.35) (2.03) (3.15) (3.19) Other 13D filer dummy –0.005 –0.006 –0.003 –0.003 (–0.50) (–0.61) (–0.78) (–0.85) Other explicit support dummy 0.020 0.021 –0.002 –0.002 (0.98) (1.02) (–0.23) (–0.28) Hedge fund characteristics Order quartile –0.001 –0.001 –0.002 –0.003 (–0.23) (–0.37) (–1.23) (–1.53) HF prior counterresistance dummy 0.005 0.006 0.005 0.004 (0.90) (1.02) (1.02) (0.98) Firm characteristics Cash flow/assets 0.040* 0.040* –0.024 –0.024 (1.84) (1.74) (–0.93) (–0.94) log of market cap. ( \$$MM) –0.002 –0.002 0.004 0.004* (–0.71) (–0.72) (1.53) (1.75) Cash 0.007 0.006 0.018 0.022* (0.34) (0.27) (1.43) (1.90) Tobin’s q –0.005*** –0.005*** 0.000 0.000 (–3.80) (–3.88) (–0.07) (–0.07) RND 0.088 0.081 –0.021 –0.032 (1.25) (1.20) (–0.34) (–0.54) Dividend yield –0.023 –0.011 0.015 0.042 (–0.11) (–0.05) (0.10) (0.31) Leverage ratio 0.015 0.016 0.018*** 0.017*** (1.26) (1.32) (7.75) (7.66) HHI –0.005 –0.009 –0.010 –0.010 (–0.18) (–0.33) (–0.45) (–0.48) Liquidity 0.011 0.010 0.000 –0.001 (1.47) (1.47) (–0.25) (–0.61) Institutional ownership 0.003 0.003 –0.008 –0.013 (0.22) (0.23) (–0.87) (–1.33) Insider ownership 0.010 0.007 0.006 –0.002 (0.50) (0.32) (0.68) (–0.21) Pill in force dummy –0.009 –0.009 0.007 –0.024 (–1.50) (–1.42) (1.11) (–0.94) b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}$$ 0.015 –0.039*** (1.24) (–3.75) b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}+$$ b$$_{\mathrm{3}}$$ –0.008 –0.036*** (–0.93) (–2.49) N 789 789 599 599 Includes year & industry dummies? Yes Yes Yes Yes Adjusted R$$^{\mathrm{2}}$$ 0.085 0.091 0.008 0.023 This table reports OLS regressions in which the dependent variable is Activism CAR$$_{\mathrm{[-1,+1]}}$$ or Resistance CAR$$_{\mathrm{[-1,+1]}}$$. We exclude campaigns with confounding events around resistance or for which we do not directly observe hostile target resistance. For campaigns with no target firm resistance, we estimate placebo resistance CARs using the date 14 days after activism announcement as the resistance date. The appendix defines the dependent and the control variables. Any resist is set to 1 for the 356 campaigns with moderate or hostile target resistance and 0 otherwise. Hostile resist is set to 1 for the subset of 207 hostile target resistance campaigns with or without formal hedge counterresistance and 0 otherwise. Opposed hostile resist is set to 1 for the subset of 138 hostile target resistance campaigns with formal hedge fund counterresistance and 0 otherwise. The missing dummy is for the 465 campaigns with no observed resistance. Regressions include year and 1-digit SIC industry dummies. Standard errors are clustered by time and hedge fund, and t-values are reported below the coefficients in parentheses. HHI, Herfindahl-Hirschman index. ***, **, and * denotes statistical significance for the 1%, 5%, and 10% levels, respectively. Table 4 Short-term market reaction to activism and resistance Activism CAR$$_{\mathrm{[-1,+1]}}$$ Resistance CAR$$_{\mathrm{[-1,+1]}}$$ (1) (2) (3) (4) Resistance categorical variables Any resist (b$$_{\mathrm{1}})$$ 0.007 0.016* –0.013*** –0.001 (1.37) (1.65) (–2.98) (–0.14) Hostile resist (b$$_{\mathrm{2}})$$ –0.001 –0.038*** (–0.07) (–3.10) Opposed hostile resist (b$$_{\mathrm{3}})$$ –0.024 0.003 (–1.60) (0.28) Campaign characteristics Hedge fund offer dummy 0.076*** 0.075*** 0.020* 0.024** (4.64) (4.39) (1.76) (2.13) Other takeover-related dummy 0.019** 0.019** –0.005 –0.005 (1.97) (2.08) (–0.74) (–0.71) Percentage owned 0.000 0.000 –0.001* 0.000 (0.16) (0.24) (–1.74) (–1.11) Hedge fund in SEC group dummy 0.036** 0.034** 0.019*** 0.021*** (2.35) (2.03) (3.15) (3.19) Other 13D filer dummy –0.005 –0.006 –0.003 –0.003 (–0.50) (–0.61) (–0.78) (–0.85) Other explicit support dummy 0.020 0.021 –0.002 –0.002 (0.98) (1.02) (–0.23) (–0.28) Hedge fund characteristics Order quartile –0.001 –0.001 –0.002 –0.003 (–0.23) (–0.37) (–1.23) (–1.53) HF prior counterresistance dummy 0.005 0.006 0.005 0.004 (0.90) (1.02) (1.02) (0.98) Firm characteristics Cash flow/assets 0.040* 0.040* –0.024 –0.024 (1.84) (1.74) (–0.93) (–0.94) log of market cap. ( $\$$MM) –0.002 –0.002 0.004 0.004* (–0.71) (–0.72) (1.53) (1.75) Cash 0.007 0.006 0.018 0.022* (0.34) (0.27) (1.43) (1.90) Tobin’s q –0.005*** –0.005*** 0.000 0.000 (–3.80) (–3.88) (–0.07) (–0.07) RND 0.088 0.081 –0.021 –0.032 (1.25) (1.20) (–0.34) (–0.54) Dividend yield –0.023 –0.011 0.015 0.042 (–0.11) (–0.05) (0.10) (0.31) Leverage ratio 0.015 0.016 0.018*** 0.017*** (1.26) (1.32) (7.75) (7.66) HHI –0.005 –0.009 –0.010 –0.010 (–0.18) (–0.33) (–0.45) (–0.48) Liquidity 0.011 0.010 0.000 –0.001 (1.47) (1.47) (–0.25) (–0.61) Institutional ownership 0.003 0.003 –0.008 –0.013 (0.22) (0.23) (–0.87) (–1.33) Insider ownership 0.010 0.007 0.006 –0.002 (0.50) (0.32) (0.68) (–0.21) Pill in force dummy –0.009 –0.009 0.007 –0.024 (–1.50) (–1.42) (1.11) (–0.94) b$$_{\mathrm{1}}+$$b$$_{\mathrm{2}}$$0.015 –0.039*** (1.24) (–3.75) b$$_{\mathrm{1}}+$$b$$_{\mathrm{2}}+$$b$$_{\mathrm{3}}$$–0.008 –0.036*** (–0.93) (–2.49) N 789 789 599 599 Includes year & industry dummies? Yes Yes Yes Yes Adjusted R$$^{\mathrm{2}}$$0.085 0.091 0.008 0.023 Activism CAR$$_{\mathrm{[-1,+1]}}$$Resistance CAR$$_{\mathrm{[-1,+1]}}$$(1) (2) (3) (4) Resistance categorical variables Any resist (b$$_{\mathrm{1}})$$0.007 0.016* –0.013*** –0.001 (1.37) (1.65) (–2.98) (–0.14) Hostile resist (b$$_{\mathrm{2}})$$–0.001 –0.038*** (–0.07) (–3.10) Opposed hostile resist (b$$_{\mathrm{3}})$$–0.024 0.003 (–1.60) (0.28) Campaign characteristics Hedge fund offer dummy 0.076*** 0.075*** 0.020* 0.024** (4.64) (4.39) (1.76) (2.13) Other takeover-related dummy 0.019** 0.019** –0.005 –0.005 (1.97) (2.08) (–0.74) (–0.71) Percentage owned 0.000 0.000 –0.001* 0.000 (0.16) (0.24) (–1.74) (–1.11) Hedge fund in SEC group dummy 0.036** 0.034** 0.019*** 0.021*** (2.35) (2.03) (3.15) (3.19) Other 13D filer dummy –0.005 –0.006 –0.003 –0.003 (–0.50) (–0.61) (–0.78) (–0.85) Other explicit support dummy 0.020 0.021 –0.002 –0.002 (0.98) (1.02) (–0.23) (–0.28) Hedge fund characteristics Order quartile –0.001 –0.001 –0.002 –0.003 (–0.23) (–0.37) (–1.23) (–1.53) HF prior counterresistance dummy 0.005 0.006 0.005 0.004 (0.90) (1.02) (1.02) (0.98) Firm characteristics Cash flow/assets 0.040* 0.040* –0.024 –0.024 (1.84) (1.74) (–0.93) (–0.94) log of market cap. ( \$$MM) –0.002 –0.002 0.004 0.004* (–0.71) (–0.72) (1.53) (1.75) Cash 0.007 0.006 0.018 0.022* (0.34) (0.27) (1.43) (1.90) Tobin’s q –0.005*** –0.005*** 0.000 0.000 (–3.80) (–3.88) (–0.07) (–0.07) RND 0.088 0.081 –0.021 –0.032 (1.25) (1.20) (–0.34) (–0.54) Dividend yield –0.023 –0.011 0.015 0.042 (–0.11) (–0.05) (0.10) (0.31) Leverage ratio 0.015 0.016 0.018*** 0.017*** (1.26) (1.32) (7.75) (7.66) HHI –0.005 –0.009 –0.010 –0.010 (–0.18) (–0.33) (–0.45) (–0.48) Liquidity 0.011 0.010 0.000 –0.001 (1.47) (1.47) (–0.25) (–0.61) Institutional ownership 0.003 0.003 –0.008 –0.013 (0.22) (0.23) (–0.87) (–1.33) Insider ownership 0.010 0.007 0.006 –0.002 (0.50) (0.32) (0.68) (–0.21) Pill in force dummy –0.009 –0.009 0.007 –0.024 (–1.50) (–1.42) (1.11) (–0.94) b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}$$ 0.015 –0.039*** (1.24) (–3.75) b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}+$$ b$$_{\mathrm{3}}$$ –0.008 –0.036*** (–0.93) (–2.49) N 789 789 599 599 Includes year & industry dummies? Yes Yes Yes Yes Adjusted R$$^{\mathrm{2}}$$ 0.085 0.091 0.008 0.023 This table reports OLS regressions in which the dependent variable is Activism CAR$$_{\mathrm{[-1,+1]}}$$ or Resistance CAR$$_{\mathrm{[-1,+1]}}$$. We exclude campaigns with confounding events around resistance or for which we do not directly observe hostile target resistance. For campaigns with no target firm resistance, we estimate placebo resistance CARs using the date 14 days after activism announcement as the resistance date. The appendix defines the dependent and the control variables. Any resist is set to 1 for the 356 campaigns with moderate or hostile target resistance and 0 otherwise. Hostile resist is set to 1 for the subset of 207 hostile target resistance campaigns with or without formal hedge counterresistance and 0 otherwise. Opposed hostile resist is set to 1 for the subset of 138 hostile target resistance campaigns with formal hedge fund counterresistance and 0 otherwise. The missing dummy is for the 465 campaigns with no observed resistance. Regressions include year and 1-digit SIC industry dummies. Standard errors are clustered by time and hedge fund, and t-values are reported below the coefficients in parentheses. HHI, Herfindahl-Hirschman index. ***, **, and * denotes statistical significance for the 1%, 5%, and 10% levels, respectively. Column 2, which includes all three resistance variables, indicates that campaigns with moderate resistance have marginally better CARs (by 1.6%) relative to campaigns with no resistance. The other resistance categorical coefficients are never significant. Other control variables have magnitudes and significance similar to those in Column 1. These results provide little evidence of systematic differences in activism announcement returns across the resistance subsamples. Taken together, these results indicate that the market predicts neither target firm resistance nor hedge fund counterresistance at activism onset. 3.2 Short-term target firm resistance announcement returns Columns 3 and 4 of Table 4 analyze the short-term market reaction (3-day CAR) to the resistance announcement using OLS regressions. The univariate results in Table 1 find that CARs for moderate resistance are a statistically insignificant 0.70%, whereas CARs for hostile resistance are negative and significant at $$-$$2.70% for unopposed hostile resistance and $$-$$2.00% for opposed hostile resistance. We calculate resistance CARs for all campaigns in which we observe moderate or hostile target resistance. The campaigns for which we infer, but do not observe, hostile resistance are excluded from these regressions. We also exclude campaigns with confounding events around resistance. For campaigns with no target firm resistance, we calculate placebo resistance CARs around the date occurring 14 days after activism, since 14 days is the average time between the activism announcement and the target firm resistance announcement. Regressions include year and one-digit SIC industry dummies; standard errors are clustered by year and hedge fund; and t-values are reported in parentheses. Results from Column 3 indicate that resistance announcement returns for campaigns with any resistance are significantly worse than placebo resistance announcement returns for campaigns with no target resistance. Resistance CARs are higher when the hedge fund makes an offer or is part of an SEC-defined group and are lower when the hedge fund owns a higher percentage of shares. Finally, resistance CARs are much higher when target firms’ leverage is higher. Results from Column 4, which include all resistance categorical variables, indicate that the results of Column 3 are entirely driven by the subsample of hostile resistance. The market reaction to moderate resistance is an insignificant $$-$$0.1%, whereas the reaction to unopposed hostile resistance is a highly significant $$-$$3.9% relative to no resistance (b$$_{\mathrm{1}}+$$b$$_{\mathrm{2}})$$. The impact of opposed hostile resistance relative to no resistance of $$-$$3.6% ((b$$_{\mathrm{1}}+$$b$$_{\mathrm{2}}+$$b$$_{\mathrm{3}})$$ is statistically significant as well. By contrast, resistance announcement CARs for opposed hostile resistance do not differ from CARs for unopposed hostile resistance, indicating that the market does not anticipate future formal hedge fund counterresistance. Overall, results from Table 4 indicate that first, the market does not respond to the announcement of moderate target resistance, and second, the market negatively responds to the announcement of hostile target resistance whether or not it is followed by formal hedge fund counterresistance. These findings imply that the market believes that hostile target firm resistance may undo some of the expected positive effects of hedge fund activism. 3.3 Resistance and mergers Table 5 presents the results of logit regressions with the dependent variable set to 1 if the target firm merges within 18 months of activism onset and 0 otherwise, following Greenwood and Schor (2009). Columns 1 and 2 include all 821 activism campaigns. The number of observations is reduced to 819 because of two cases in which a year dummy perfectly predicts the dependent variable. Columns 3 and 4 include only the subsample of campaigns in which the hedge fund makes an offer or suggests that the firm sell itself. Instead of regression coefficients, the table reports the probability change for a move from 0 to 1 for indicator variables and from the 25th percentile to the 75th percentile for continuous variables, with all other variables set to their means. Regressions include year and one-digit SIC industry dummies; standard errors are clustered by year and hedge fund; and z-values are reported in parentheses. As a benchmark, the unconditional probability of a merger is 18% for the campaigns in Columns 1 and 2 and 28% for campaigns in Columns 3 and 4. Table 5 Resistance and mergers All campaigns, firm merges within 18 months of activism Takeover-related campaigns, firm merges within 18 months of activism (1) (2) (3) (4) Resistance categorical variables Any resist (b$$_{\mathrm{1}})$$ –0.008 –0.013 –0.090 –0.085 (–0.32) (–0.40) (–1.42) (–1.00) Hostile resist (b$$_{\mathrm{2}})$$ –0.037 –0.189*** (–1.04) (–2.68) Opposed hostile resist (b$$_{\mathrm{3}})$$ 0.076* 0.364** (1.77) (2.48) Campaign characteristics Hedge fund offer dummy 0.247*** 0.252*** 0.073 0.104* (5.96) (5.56) (1.50) (1.76) Other takeover dummy 0.163*** 0.161*** (4.25) (4.25) Percentage owned 0.015 0.016 0.036 0.034 (1.20) (1.39) (0.99) (0.86) Hedge fund in SEC group dummy 0.082 0.088 0.238* 0.026** (1.24) (1.35) (1.90) (1.98) Other 13D filer dummy –0.019 –0.018 –0.012 –0.006 (–0.40) (–0.38) (–0.14) (–0.08) Other explicit support dummy 0.579** 0.580** 0.509*** 0.516*** (2.24) (2.26) (4.23) (3.99) Hedge fund characteristics Order quartile 0.042 0.043 0.167 0.170 (1.21) (1.23) (1.62) (1.61) HF prior counterresistance dummy 0.007 0.004 0.011 0.004 (0.21) (0.12) (0.18) (0.05) Firm characteristics Cash flow/assets 0.014 0.014 0.026 0.033* (1.30) (1.34) (1.36) (1.87) log of market cap. ($\$$MM) –0.025 –0.024 0.014 0.011 (–1.07) (–1.03) (0.29) (0.21) Cash 0.014 0.016 0.025 0.037 (0.75) (0.96) (0.37) (0.60) Tobin’s q –0.003 –0.004 –0.035** –0.038** (–0.41) (–0.48) (–2.07) (–2.03) RND 0.015 0.016 0.053** 0.060** (0.76) (0.79) (1.98) (2.23) Dividend yield –0.009 –0.008 0.013 0.013 (–1.59) (–1.63) (1.08) (1.13) Leverage 0.001 0.002 –0.014 –0.005 (0.05) (0.05) (–0.21) (–0.07) HHI –0.014 –0.012 –0.027 –0.019 (–0.87) (–0.78) (–1.19) (–1.04) Liquidity 0.017*** 0.017*** 0.022* 0.020* (2.76) (2.83) (1.77) (1.67) Institutional ownership 0.014 0.013 0.005 –0.000 (0.86) (0.73) (0.15) (0.00) Insider ownership –0.001 –0.000 –0.046 –0.050 (–0.04) (–0.05) (–1.59) (–1.64) Pill in force dummy –0.009 –0.012 –0.012 –0.033 (–0.54) (–0.65) (–0.30) (–0.90) b$$_{\mathrm{1}}+$$b$$_{\mathrm{2}}$$–0.055 –0.274*** (–1.21) (–2.62) b$$_{\mathrm{1}}+$$b$$_{\mathrm{2}}+$$b$$_{\mathrm{3}}$$0.021 0.090 (0.54) (0.41) N 819 819 331 331 Includes year & industry dummies? Yes Yes Yes Yes Pseudo R$$^{\mathrm{2}}$$0.135 0.137 0.194 0.210 All campaigns, firm merges within 18 months of activism Takeover-related campaigns, firm merges within 18 months of activism (1) (2) (3) (4) Resistance categorical variables Any resist (b$$_{\mathrm{1}})$$–0.008 –0.013 –0.090 –0.085 (–0.32) (–0.40) (–1.42) (–1.00) Hostile resist (b$$_{\mathrm{2}})$$–0.037 –0.189*** (–1.04) (–2.68) Opposed hostile resist (b$$_{\mathrm{3}})$$0.076* 0.364** (1.77) (2.48) Campaign characteristics Hedge fund offer dummy 0.247*** 0.252*** 0.073 0.104* (5.96) (5.56) (1.50) (1.76) Other takeover dummy 0.163*** 0.161*** (4.25) (4.25) Percentage owned 0.015 0.016 0.036 0.034 (1.20) (1.39) (0.99) (0.86) Hedge fund in SEC group dummy 0.082 0.088 0.238* 0.026** (1.24) (1.35) (1.90) (1.98) Other 13D filer dummy –0.019 –0.018 –0.012 –0.006 (–0.40) (–0.38) (–0.14) (–0.08) Other explicit support dummy 0.579** 0.580** 0.509*** 0.516*** (2.24) (2.26) (4.23) (3.99) Hedge fund characteristics Order quartile 0.042 0.043 0.167 0.170 (1.21) (1.23) (1.62) (1.61) HF prior counterresistance dummy 0.007 0.004 0.011 0.004 (0.21) (0.12) (0.18) (0.05) Firm characteristics Cash flow/assets 0.014 0.014 0.026 0.033* (1.30) (1.34) (1.36) (1.87) log of market cap. ( \$$MM) –0.025 –0.024 0.014 0.011 (–1.07) (–1.03) (0.29) (0.21) Cash 0.014 0.016 0.025 0.037 (0.75) (0.96) (0.37) (0.60) Tobin’s q –0.003 –0.004 –0.035** –0.038** (–0.41) (–0.48) (–2.07) (–2.03) RND 0.015 0.016 0.053** 0.060** (0.76) (0.79) (1.98) (2.23) Dividend yield –0.009 –0.008 0.013 0.013 (–1.59) (–1.63) (1.08) (1.13) Leverage 0.001 0.002 –0.014 –0.005 (0.05) (0.05) (–0.21) (–0.07) HHI –0.014 –0.012 –0.027 –0.019 (–0.87) (–0.78) (–1.19) (–1.04) Liquidity 0.017*** 0.017*** 0.022* 0.020* (2.76) (2.83) (1.77) (1.67) Institutional ownership 0.014 0.013 0.005 –0.000 (0.86) (0.73) (0.15) (0.00) Insider ownership –0.001 –0.000 –0.046 –0.050 (–0.04) (–0.05) (–1.59) (–1.64) Pill in force dummy –0.009 –0.012 –0.012 –0.033 (–0.54) (–0.65) (–0.30) (–0.90) b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}$$ –0.055 –0.274*** (–1.21) (–2.62) b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}+$$ b$$_{\mathrm{3}}$$ 0.021 0.090 (0.54) (0.41) N 819 819 331 331 Includes year & industry dummies? Yes Yes Yes Yes Pseudo R$$^{\mathrm{2}}$$ 0.135 0.137 0.194 0.210 This table reports logit analyses in which the dependent variable is a dummy set to 1 if the target firm merges within 18 months of activism onset and 0 otherwise. The table reports marginal effects. Columns 1 and 2 include all activism campaigns. Columns 3 and 4 include only takeover-related campaigns. Any resist is set to 1 for the 356 campaigns with moderate or hostile target resistance and 0 otherwise. Hostile resist is set to 1 for the subset of 207 hostile target resistance campaigns with or without formal hedge counterresistance and 0 otherwise. Opposed hostile resist is set to 1 for the subset of 138 hostile target resistance campaigns with formal hedge fund counterresistance and 0 otherwise. The reported values represent the probability change for a move from 0 to 1 for indicator variables and from the 25th to the 75th percentile for continuous variables, setting all other variables at their means. All specifications include year and 1-digit SIC industry dummies. Standard errors are clustered by time and hedge fund, and z-values are reported below the coefficients in parentheses. HHI, Herfindahl-Hirschman index. ***, **, and * denotes statistical significance for the 1%, 5%, and 10% levels, respectively. Table 5 Resistance and mergers All campaigns, firm merges within 18 months of activism Takeover-related campaigns, firm merges within 18 months of activism (1) (2) (3) (4) Resistance categorical variables Any resist (b$$_{\mathrm{1}})$$ –0.008 –0.013 –0.090 –0.085 (–0.32) (–0.40) (–1.42) (–1.00) Hostile resist (b$$_{\mathrm{2}})$$ –0.037 –0.189*** (–1.04) (–2.68) Opposed hostile resist (b$$_{\mathrm{3}})$$ 0.076* 0.364** (1.77) (2.48) Campaign characteristics Hedge fund offer dummy 0.247*** 0.252*** 0.073 0.104* (5.96) (5.56) (1.50) (1.76) Other takeover dummy 0.163*** 0.161*** (4.25) (4.25) Percentage owned 0.015 0.016 0.036 0.034 (1.20) (1.39) (0.99) (0.86) Hedge fund in SEC group dummy 0.082 0.088 0.238* 0.026** (1.24) (1.35) (1.90) (1.98) Other 13D filer dummy –0.019 –0.018 –0.012 –0.006 (–0.40) (–0.38) (–0.14) (–0.08) Other explicit support dummy 0.579** 0.580** 0.509*** 0.516*** (2.24) (2.26) (4.23) (3.99) Hedge fund characteristics Order quartile 0.042 0.043 0.167 0.170 (1.21) (1.23) (1.62) (1.61) HF prior counterresistance dummy 0.007 0.004 0.011 0.004 (0.21) (0.12) (0.18) (0.05) Firm characteristics Cash flow/assets 0.014 0.014 0.026 0.033* (1.30) (1.34) (1.36) (1.87) log of market cap. ( $\$$MM) –0.025 –0.024 0.014 0.011 (–1.07) (–1.03) (0.29) (0.21) Cash 0.014 0.016 0.025 0.037 (0.75) (0.96) (0.37) (0.60) Tobin’s q –0.003 –0.004 –0.035** –0.038** (–0.41) (–0.48) (–2.07) (–2.03) RND 0.015 0.016 0.053** 0.060** (0.76) (0.79) (1.98) (2.23) Dividend yield –0.009 –0.008 0.013 0.013 (–1.59) (–1.63) (1.08) (1.13) Leverage 0.001 0.002 –0.014 –0.005 (0.05) (0.05) (–0.21) (–0.07) HHI –0.014 –0.012 –0.027 –0.019 (–0.87) (–0.78) (–1.19) (–1.04) Liquidity 0.017*** 0.017*** 0.022* 0.020* (2.76) (2.83) (1.77) (1.67) Institutional ownership 0.014 0.013 0.005 –0.000 (0.86) (0.73) (0.15) (0.00) Insider ownership –0.001 –0.000 –0.046 –0.050 (–0.04) (–0.05) (–1.59) (–1.64) Pill in force dummy –0.009 –0.012 –0.012 –0.033 (–0.54) (–0.65) (–0.30) (–0.90) b$$_{\mathrm{1}}+$$b$$_{\mathrm{2}}$$–0.055 –0.274*** (–1.21) (–2.62) b$$_{\mathrm{1}}+$$b$$_{\mathrm{2}}+$$b$$_{\mathrm{3}}$$0.021 0.090 (0.54) (0.41) N 819 819 331 331 Includes year & industry dummies? Yes Yes Yes Yes Pseudo R$$^{\mathrm{2}}$$0.135 0.137 0.194 0.210 All campaigns, firm merges within 18 months of activism Takeover-related campaigns, firm merges within 18 months of activism (1) (2) (3) (4) Resistance categorical variables Any resist (b$$_{\mathrm{1}})$$–0.008 –0.013 –0.090 –0.085 (–0.32) (–0.40) (–1.42) (–1.00) Hostile resist (b$$_{\mathrm{2}})$$–0.037 –0.189*** (–1.04) (–2.68) Opposed hostile resist (b$$_{\mathrm{3}})$$0.076* 0.364** (1.77) (2.48) Campaign characteristics Hedge fund offer dummy 0.247*** 0.252*** 0.073 0.104* (5.96) (5.56) (1.50) (1.76) Other takeover dummy 0.163*** 0.161*** (4.25) (4.25) Percentage owned 0.015 0.016 0.036 0.034 (1.20) (1.39) (0.99) (0.86) Hedge fund in SEC group dummy 0.082 0.088 0.238* 0.026** (1.24) (1.35) (1.90) (1.98) Other 13D filer dummy –0.019 –0.018 –0.012 –0.006 (–0.40) (–0.38) (–0.14) (–0.08) Other explicit support dummy 0.579** 0.580** 0.509*** 0.516*** (2.24) (2.26) (4.23) (3.99) Hedge fund characteristics Order quartile 0.042 0.043 0.167 0.170 (1.21) (1.23) (1.62) (1.61) HF prior counterresistance dummy 0.007 0.004 0.011 0.004 (0.21) (0.12) (0.18) (0.05) Firm characteristics Cash flow/assets 0.014 0.014 0.026 0.033* (1.30) (1.34) (1.36) (1.87) log of market cap. ( \$$MM) –0.025 –0.024 0.014 0.011 (–1.07) (–1.03) (0.29) (0.21) Cash 0.014 0.016 0.025 0.037 (0.75) (0.96) (0.37) (0.60) Tobin’s q –0.003 –0.004 –0.035** –0.038** (–0.41) (–0.48) (–2.07) (–2.03) RND 0.015 0.016 0.053** 0.060** (0.76) (0.79) (1.98) (2.23) Dividend yield –0.009 –0.008 0.013 0.013 (–1.59) (–1.63) (1.08) (1.13) Leverage 0.001 0.002 –0.014 –0.005 (0.05) (0.05) (–0.21) (–0.07) HHI –0.014 –0.012 –0.027 –0.019 (–0.87) (–0.78) (–1.19) (–1.04) Liquidity 0.017*** 0.017*** 0.022* 0.020* (2.76) (2.83) (1.77) (1.67) Institutional ownership 0.014 0.013 0.005 –0.000 (0.86) (0.73) (0.15) (0.00) Insider ownership –0.001 –0.000 –0.046 –0.050 (–0.04) (–0.05) (–1.59) (–1.64) Pill in force dummy –0.009 –0.012 –0.012 –0.033 (–0.54) (–0.65) (–0.30) (–0.90) b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}$$ –0.055 –0.274*** (–1.21) (–2.62) b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}+$$ b$$_{\mathrm{3}}$$ 0.021 0.090 (0.54) (0.41) N 819 819 331 331 Includes year & industry dummies? Yes Yes Yes Yes Pseudo R$$^{\mathrm{2}}$$ 0.135 0.137 0.194 0.210 This table reports logit analyses in which the dependent variable is a dummy set to 1 if the target firm merges within 18 months of activism onset and 0 otherwise. The table reports marginal effects. Columns 1 and 2 include all activism campaigns. Columns 3 and 4 include only takeover-related campaigns. Any resist is set to 1 for the 356 campaigns with moderate or hostile target resistance and 0 otherwise. Hostile resist is set to 1 for the subset of 207 hostile target resistance campaigns with or without formal hedge counterresistance and 0 otherwise. Opposed hostile resist is set to 1 for the subset of 138 hostile target resistance campaigns with formal hedge fund counterresistance and 0 otherwise. The reported values represent the probability change for a move from 0 to 1 for indicator variables and from the 25th to the 75th percentile for continuous variables, setting all other variables at their means. All specifications include year and 1-digit SIC industry dummies. Standard errors are clustered by time and hedge fund, and z-values are reported below the coefficients in parentheses. HHI, Herfindahl-Hirschman index. ***, **, and * denotes statistical significance for the 1%, 5%, and 10% levels, respectively. Column 1, which includes the any resist categorical variable, indicates that resistance in general does not affect the likelihood that a firm will merge. Column 2 includes all three resistance categorical variables and shows that neither moderate nor hostile resistance has a significant impact on merger probability. However, the incremental effect of formal counterresistance is significantly positive, increasing the merger probability by a relative 42% (0.076/0.18) compared to campaigns with unopposed hostile resistance. However, the impact of opposed hostile resistance relative to no resistance (b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}+$$ b$$_{\mathrm{3}})$$ is insignificant. The probability of a merger is 140% higher (0.252/0.18) when the hedge fund makes an offer and 89% higher (0.161/0.18) when the hedge fund states a goal of a merger. Further, the likelihood of a merger is 322% higher (0.580/0.18) when other activists state their explicit support for the hedge fund activist, consistent with Brav, Dasgupta, and Mathews (2017) and Becht et al. (2017), who argue that activists play a catalytic role in coordinating the actions of other institutional investors. The importance of coordinated shareholder actions is particularly relevant here, since mergers require a shareholder vote. Finally, more liquid firms are more likely than less liquid firms to merge. These results together with the results from Table 2 lead to concern about a potential selection bias because takeover-related campaigns are more likely to result in acquisitions and are more likely to incur hostile resistance. As a partial control for this bias, we repeat the merger regressions, including only the subset of takeover-related activism campaigns. This subset includes campaigns for which either the hedge fund offer dummy or the other takeover-related dummy equals one. Results for Column 3 are similar to Column 1: there is no relation between the any resist dummy variable and the probability of a merger. However, Column 4 indicates that the incremental effect of hostility is to decrease the relative probability of a merger by 68% (0.189/0.28). Summing the first two coefficients, the relative probability of a merger is 98% (0.274/0.28) lower for campaigns with unopposed hostile resistance relative to campaigns with no resistance. The incremental impact of formal counterresistance by the hedge fund is a highly significant 130% (0.364/0.28). Summing all three resistance coefficients, the likelihood of a merger for campaigns with opposed hostile resistance does not significantly differ from the likelihood for campaigns with no resistance. These analyses provide strong evidence that (1) moderate resistance has no impact on the likelihood of a merger, (2) unopposed hostile resistance significantly negatively affects the probability of a merger, (3) formal counterresistance appears to reverse the negative effect of hostile target firm resistance on merger likelihood, and (4) hedge fund collaboration significantly increases merger likelihood. These results support our conjecture that entrenched target managers are more likely to engage in hostile resistance when they feel most threatened by a hedge fund offer or by highly concentrated institutional ownership. 3.4 Operating performance and resistance Table 6 presents results of OLS regressions in which the dependent variable is either the change in cash flow/assets or ROA measured 1 or 2 years after activism announcement, relative to the year before activism. Univariate results in Table 1 indicate significant differences in changes in cash flows among the subsamples; for example, cash flow changes are much better for campaigns with formal hedge fund counterresistance relative to campaigns without formal hedge fund counterresistance. In addition to other control variables, the regressions include 1-year lagged realizations of operating performance. Regressions include all activist campaigns in which the target firm is extant in the 1 or 2 years following activism. Regressions include year and one-digit SIC industry dummies; standard errors are clustered by year and hedge fund; and t-values are reported in parentheses. Table 6 Resistance and operating performance $$\Delta$$ in CF 1 yr $$\Delta$$ in CF 2 yr $$\Delta$$ in ROA 1 yr $$\Delta$$ in ROA 2 yr (1) (2) (3) (4) (5) (6) (7) (8) Resistance categorical variables Any resist (b$$_{\mathrm{1}})$$ 0.003 0.010 0.011 0.022** –0.003 –0.000 0.014 0.018 (0.22) (0.79) (1.16) (2.23) (–0.34) (–0.05) (1.36) (1.51) Hostile resist (b$$_{\mathrm{2}})$$ –0.065*** –0.050* –0.026* –0.032 (–2.68) (–1.85) (–1.67) (–1.26) Opposed hostile resist (b$$_{\mathrm{3}})$$ 0.077** 0.046** 0.032 0.038* (2.18) (2.41) (1.17) (1.84) Campaign characteristics Hedge fund offer dummy 0.005 0.010 –0.007 –0.003 0.012 0.014 –0.010 –0.008 (0.22) (0.48) (–0.38) (–0.20) (0.54) (0.63) (–0.43) (–0.34) Other takeover dummy –0.010 –0.008 –0.003 –0.001 0.002 0.003 –0.014 –0.013 (–1.09) (–0.93) (–0.24) (–0.10) (0.25) (0.30) (–1.43) (–1.32) Percentage owned –0.001 0.000 –0.004** –0.003** 0.001 0.001 –0.002* –0.002 (–0.56) (–0.15) (–2.26) (–2.07) (0.89) (1.04) (–1.74) (–1.57) Hedge fund in SEC group dummy 0.007 0.011 –0.037 –0.035 0.003 0.005 –0.044 –0.042 (0.36) (0.74) (–0.95) (–0.90) (0.15) (0.29) (–1.40) (–1.33) Other 13D filer dummy –0.014 –0.011 –0.005 –0.003 –0.014 –0.013 –0.000 0.002 (–1.00) (–0.79) (–0.68) (–0.38) (–1.53) (–1.38) (–0.02) (0.15) Other explicit support dummy 0.034 0.037 0.043** 0.044** 0.033** 0.034** 0.007 0.007 (1.53) (1.65) (2.31) (2.27) (2.33) (2.46) (0.32) (0.34) Hedge fund characteristics Order quartile –0.009 –0.010 0.004 0.003 –0.006 –0.007 0.001 –0.000 (–1.33) (–1.52) (0.71) (0.44) (–1.25) (–1.41) (0.18) (–0.02) HF prior counterresistance dummy 0.046*** 0.046*** 0.022 0.023 0.028** 0.028** 0.026 0.026 (3.93) (3.86) (1.04) (1.08) (2.38) (2.47) (1.52) (1.52) Firm characteristics Cash flow/assets or ROA/assets –0.469*** –0.468*** –0.701*** –0.698*** –0.453*** –0.455*** –0.490*** –0.492*** (–4.86) (–4.93) (–9.97) (–9.68) (–6.60) (–6.65) (–6.62) (–6.64) log of market cap. ($\$$MM) 0.009 0.010* 0.005 0.005 0.003 0.004 0.002 0.003 (1.61) (1.79) (1.05) (1.15) (0.85) (0.96) (0.65) (0.81) Cash –0.152*** –0.142*** –0.169*** –0.164*** –0.147*** –0.144*** –0.090*** –0.087*** (–4.35) (–3.48) (–3.17) (–3.00) (–7.26) (–6.65) (–3.22) (–3.02) Tobin’s q 0.003 0.002 0.004 0.003 0.001 0.000 0.005 0.005 (0.43) (0.31) (0.70) (0.63) (0.15) (0.05) (0.92) (0.80) RND –0.050 –0.055 –0.145 –0.145 –0.141 –0.145 –0.174* –0.175* (–0.29) (–0.33) (–1.15) (–1.17) (–1.32) (–1.37) (–1.76) (–1.79) Dividend yield –0.376 –0.401 –0.111 –0.114 –0.095 –0.106 –0.174 –0.187 (–1.41) (–1.48) (–0.54) (–0.61) (–0.64) (–0.67) (–1.09) (–1.09) Leverage –0.007 –0.010 –0.034 –0.036 0.005 0.003 –0.014 –0.016 (–0.37) (–0.63) (–1.20) (–1.26) (0.28) (0.19) (–0.71) (–0.79) HHI –0.048 –0.037 –0.036 –0.031 –0.066* –0.061* –0.048* –0.044* (–0.85) (–0.67) (–1.11) (–1.00) (–1.87) (–1.75) (–1.80) (–1.67) Liquidity 0.009 0.009 0.007 0.007 0.012 0.013 0.013* 0.013* (0.68) (0.75) (1.28) (1.66) (1.03) (1.07) (1.76) (1.93) Institutional ownership 0.032 0.024 0.024 0.019 0.038 0.035 0.030 0.027 (0.51) (0.39) (0.58) (0.45) (0.95) (0.84) (0.79) (0.68) Insider ownership –0.024 –0.027 –0.009 –0.011 –0.039 –0.040 –0.007 –0.007 (–0.52) (–0.56) (–0.19) (–0.22) (–1.11) (–1.12) (–0.22) (–0.22) Pill in force dummy –0.005 –0.010 –0.002 –0.006 0.002 –0.000 –0.000 –0.003 (–0.31) (–0.65) (–0.13) (–0.38) (0.35) (–0.02) (–0.02) (–0.34) b$$_{\mathrm{1 }}+$$b$$_{\mathrm{2}}$$–0.055** –0.028 –0.027 –0.014 (–2.03) (–1.55) (–1.59) (–0.76) b$$_{\mathrm{1}}+$$b$$_{\mathrm{2}}+$$b$$_{\mathrm{3}}$$0.022 0.017 0.005 0.024 (0.99) (0.66) (0.26) (1.05) N 629 629 555 555 629 629 557 557 Includes year & industry dummies? Yes Yes Yes Yes Yes Yes Yes Yes Adjusted R$$^{\mathrm{2}}$$0.166 0.175 0.323 0.332 0.240 0.242 0.237 0.239$$\Delta $$in CF 1 yr$$\Delta $$in CF 2 yr$$\Delta $$in ROA 1 yr$$\Delta $$in ROA 2 yr (1) (2) (3) (4) (5) (6) (7) (8) Resistance categorical variables Any resist (b$$_{\mathrm{1}})$$0.003 0.010 0.011 0.022** –0.003 –0.000 0.014 0.018 (0.22) (0.79) (1.16) (2.23) (–0.34) (–0.05) (1.36) (1.51) Hostile resist (b$$_{\mathrm{2}})$$–0.065*** –0.050* –0.026* –0.032 (–2.68) (–1.85) (–1.67) (–1.26) Opposed hostile resist (b$$_{\mathrm{3}})$$0.077** 0.046** 0.032 0.038* (2.18) (2.41) (1.17) (1.84) Campaign characteristics Hedge fund offer dummy 0.005 0.010 –0.007 –0.003 0.012 0.014 –0.010 –0.008 (0.22) (0.48) (–0.38) (–0.20) (0.54) (0.63) (–0.43) (–0.34) Other takeover dummy –0.010 –0.008 –0.003 –0.001 0.002 0.003 –0.014 –0.013 (–1.09) (–0.93) (–0.24) (–0.10) (0.25) (0.30) (–1.43) (–1.32) Percentage owned –0.001 0.000 –0.004** –0.003** 0.001 0.001 –0.002* –0.002 (–0.56) (–0.15) (–2.26) (–2.07) (0.89) (1.04) (–1.74) (–1.57) Hedge fund in SEC group dummy 0.007 0.011 –0.037 –0.035 0.003 0.005 –0.044 –0.042 (0.36) (0.74) (–0.95) (–0.90) (0.15) (0.29) (–1.40) (–1.33) Other 13D filer dummy –0.014 –0.011 –0.005 –0.003 –0.014 –0.013 –0.000 0.002 (–1.00) (–0.79) (–0.68) (–0.38) (–1.53) (–1.38) (–0.02) (0.15) Other explicit support dummy 0.034 0.037 0.043** 0.044** 0.033** 0.034** 0.007 0.007 (1.53) (1.65) (2.31) (2.27) (2.33) (2.46) (0.32) (0.34) Hedge fund characteristics Order quartile –0.009 –0.010 0.004 0.003 –0.006 –0.007 0.001 –0.000 (–1.33) (–1.52) (0.71) (0.44) (–1.25) (–1.41) (0.18) (–0.02) HF prior counterresistance dummy 0.046*** 0.046*** 0.022 0.023 0.028** 0.028** 0.026 0.026 (3.93) (3.86) (1.04) (1.08) (2.38) (2.47) (1.52) (1.52) Firm characteristics Cash flow/assets or ROA/assets –0.469*** –0.468*** –0.701*** –0.698*** –0.453*** –0.455*** –0.490*** –0.492*** (–4.86) (–4.93) (–9.97) (–9.68) (–6.60) (–6.65) (–6.62) (–6.64) log of market cap. ( \$$MM) 0.009 0.010* 0.005 0.005 0.003 0.004 0.002 0.003 (1.61) (1.79) (1.05) (1.15) (0.85) (0.96) (0.65) (0.81) Cash –0.152*** –0.142*** –0.169*** –0.164*** –0.147*** –0.144*** –0.090*** –0.087*** (–4.35) (–3.48) (–3.17) (–3.00) (–7.26) (–6.65) (–3.22) (–3.02) Tobin’s q 0.003 0.002 0.004 0.003 0.001 0.000 0.005 0.005 (0.43) (0.31) (0.70) (0.63) (0.15) (0.05) (0.92) (0.80) RND –0.050 –0.055 –0.145 –0.145 –0.141 –0.145 –0.174* –0.175* (–0.29) (–0.33) (–1.15) (–1.17) (–1.32) (–1.37) (–1.76) (–1.79) Dividend yield –0.376 –0.401 –0.111 –0.114 –0.095 –0.106 –0.174 –0.187 (–1.41) (–1.48) (–0.54) (–0.61) (–0.64) (–0.67) (–1.09) (–1.09) Leverage –0.007 –0.010 –0.034 –0.036 0.005 0.003 –0.014 –0.016 (–0.37) (–0.63) (–1.20) (–1.26) (0.28) (0.19) (–0.71) (–0.79) HHI –0.048 –0.037 –0.036 –0.031 –0.066* –0.061* –0.048* –0.044* (–0.85) (–0.67) (–1.11) (–1.00) (–1.87) (–1.75) (–1.80) (–1.67) Liquidity 0.009 0.009 0.007 0.007 0.012 0.013 0.013* 0.013* (0.68) (0.75) (1.28) (1.66) (1.03) (1.07) (1.76) (1.93) Institutional ownership 0.032 0.024 0.024 0.019 0.038 0.035 0.030 0.027 (0.51) (0.39) (0.58) (0.45) (0.95) (0.84) (0.79) (0.68) Insider ownership –0.024 –0.027 –0.009 –0.011 –0.039 –0.040 –0.007 –0.007 (–0.52) (–0.56) (–0.19) (–0.22) (–1.11) (–1.12) (–0.22) (–0.22) Pill in force dummy –0.005 –0.010 –0.002 –0.006 0.002 –0.000 –0.000 –0.003 (–0.31) (–0.65) (–0.13) (–0.38) (0.35) (–0.02) (–0.02) (–0.34) b$$_{\mathrm{1 }}+$$ b$$_{\mathrm{2}}$$ –0.055** –0.028 –0.027 –0.014 (–2.03) (–1.55) (–1.59) (–0.76) b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}+$$ b$$_{\mathrm{3}}$$ 0.022 0.017 0.005 0.024 (0.99) (0.66) (0.26) (1.05) N 629 629 555 555 629 629 557 557 Includes year & industry dummies? Yes Yes Yes Yes Yes Yes Yes Yes Adjusted R$$^{\mathrm{2}}$$ 0.166 0.175 0.323 0.332 0.240 0.242 0.237 0.239 This table reports OLS regression results. The dependent variable is the change in cash flows or the change in ROA. The appendix defines the dependent and the control variables. Any resist is set to 1 for the 356 campaigns with moderate or hostile target resistance and 0 otherwise. Hostile resist is set to 1 for the subset of 207 hostile target resistance campaigns with or without formal hedge counterresistance and 0 otherwise. Opposed hostile resist is set to 1 for the subset of 138 hostile target resistance campaigns with formal hedge fund counterresistance and 0 otherwise. Regressions include year and 1-digit SIC industry dummies. Standard errors are clustered by time and hedge fund, and t-values are reported below the coefficients in parentheses. HHI, Herfindahl-Hirschman index***, **, and * denotes statistical significance for the 1%, 5%, and 10% levels, respectively. Table 6 Resistance and operating performance $$\Delta$$ in CF 1 yr $$\Delta$$ in CF 2 yr $$\Delta$$ in ROA 1 yr $$\Delta$$ in ROA 2 yr (1) (2) (3) (4) (5) (6) (7) (8) Resistance categorical variables Any resist (b$$_{\mathrm{1}})$$ 0.003 0.010 0.011 0.022** –0.003 –0.000 0.014 0.018 (0.22) (0.79) (1.16) (2.23) (–0.34) (–0.05) (1.36) (1.51) Hostile resist (b$$_{\mathrm{2}})$$ –0.065*** –0.050* –0.026* –0.032 (–2.68) (–1.85) (–1.67) (–1.26) Opposed hostile resist (b$$_{\mathrm{3}})$$ 0.077** 0.046** 0.032 0.038* (2.18) (2.41) (1.17) (1.84) Campaign characteristics Hedge fund offer dummy 0.005 0.010 –0.007 –0.003 0.012 0.014 –0.010 –0.008 (0.22) (0.48) (–0.38) (–0.20) (0.54) (0.63) (–0.43) (–0.34) Other takeover dummy –0.010 –0.008 –0.003 –0.001 0.002 0.003 –0.014 –0.013 (–1.09) (–0.93) (–0.24) (–0.10) (0.25) (0.30) (–1.43) (–1.32) Percentage owned –0.001 0.000 –0.004** –0.003** 0.001 0.001 –0.002* –0.002 (–0.56) (–0.15) (–2.26) (–2.07) (0.89) (1.04) (–1.74) (–1.57) Hedge fund in SEC group dummy 0.007 0.011 –0.037 –0.035 0.003 0.005 –0.044 –0.042 (0.36) (0.74) (–0.95) (–0.90) (0.15) (0.29) (–1.40) (–1.33) Other 13D filer dummy –0.014 –0.011 –0.005 –0.003 –0.014 –0.013 –0.000 0.002 (–1.00) (–0.79) (–0.68) (–0.38) (–1.53) (–1.38) (–0.02) (0.15) Other explicit support dummy 0.034 0.037 0.043** 0.044** 0.033** 0.034** 0.007 0.007 (1.53) (1.65) (2.31) (2.27) (2.33) (2.46) (0.32) (0.34) Hedge fund characteristics Order quartile –0.009 –0.010 0.004 0.003 –0.006 –0.007 0.001 –0.000 (–1.33) (–1.52) (0.71) (0.44) (–1.25) (–1.41) (0.18) (–0.02) HF prior counterresistance dummy 0.046*** 0.046*** 0.022 0.023 0.028** 0.028** 0.026 0.026 (3.93) (3.86) (1.04) (1.08) (2.38) (2.47) (1.52) (1.52) Firm characteristics Cash flow/assets or ROA/assets –0.469*** –0.468*** –0.701*** –0.698*** –0.453*** –0.455*** –0.490*** –0.492*** (–4.86) (–4.93) (–9.97) (–9.68) (–6.60) (–6.65) (–6.62) (–6.64) log of market cap. ( $\$$MM) 0.009 0.010* 0.005 0.005 0.003 0.004 0.002 0.003 (1.61) (1.79) (1.05) (1.15) (0.85) (0.96) (0.65) (0.81) Cash –0.152*** –0.142*** –0.169*** –0.164*** –0.147*** –0.144*** –0.090*** –0.087*** (–4.35) (–3.48) (–3.17) (–3.00) (–7.26) (–6.65) (–3.22) (–3.02) Tobin’s q 0.003 0.002 0.004 0.003 0.001 0.000 0.005 0.005 (0.43) (0.31) (0.70) (0.63) (0.15) (0.05) (0.92) (0.80) RND –0.050 –0.055 –0.145 –0.145 –0.141 –0.145 –0.174* –0.175* (–0.29) (–0.33) (–1.15) (–1.17) (–1.32) (–1.37) (–1.76) (–1.79) Dividend yield –0.376 –0.401 –0.111 –0.114 –0.095 –0.106 –0.174 –0.187 (–1.41) (–1.48) (–0.54) (–0.61) (–0.64) (–0.67) (–1.09) (–1.09) Leverage –0.007 –0.010 –0.034 –0.036 0.005 0.003 –0.014 –0.016 (–0.37) (–0.63) (–1.20) (–1.26) (0.28) (0.19) (–0.71) (–0.79) HHI –0.048 –0.037 –0.036 –0.031 –0.066* –0.061* –0.048* –0.044* (–0.85) (–0.67) (–1.11) (–1.00) (–1.87) (–1.75) (–1.80) (–1.67) Liquidity 0.009 0.009 0.007 0.007 0.012 0.013 0.013* 0.013* (0.68) (0.75) (1.28) (1.66) (1.03) (1.07) (1.76) (1.93) Institutional ownership 0.032 0.024 0.024 0.019 0.038 0.035 0.030 0.027 (0.51) (0.39) (0.58) (0.45) (0.95) (0.84) (0.79) (0.68) Insider ownership –0.024 –0.027 –0.009 –0.011 –0.039 –0.040 –0.007 –0.007 (–0.52) (–0.56) (–0.19) (–0.22) (–1.11) (–1.12) (–0.22) (–0.22) Pill in force dummy –0.005 –0.010 –0.002 –0.006 0.002 –0.000 –0.000 –0.003 (–0.31) (–0.65) (–0.13) (–0.38) (0.35) (–0.02) (–0.02) (–0.34) b$$_{\mathrm{1 }}+$$b$$_{\mathrm{2}}$$–0.055** –0.028 –0.027 –0.014 (–2.03) (–1.55) (–1.59) (–0.76) b$$_{\mathrm{1}}+$$b$$_{\mathrm{2}}+$$b$$_{\mathrm{3}}$$0.022 0.017 0.005 0.024 (0.99) (0.66) (0.26) (1.05) N 629 629 555 555 629 629 557 557 Includes year & industry dummies? Yes Yes Yes Yes Yes Yes Yes Yes Adjusted R$$^{\mathrm{2}}$$0.166 0.175 0.323 0.332 0.240 0.242 0.237 0.239$$\Delta $$in CF 1 yr$$\Delta $$in CF 2 yr$$\Delta $$in ROA 1 yr$$\Delta $$in ROA 2 yr (1) (2) (3) (4) (5) (6) (7) (8) Resistance categorical variables Any resist (b$$_{\mathrm{1}})$$0.003 0.010 0.011 0.022** –0.003 –0.000 0.014 0.018 (0.22) (0.79) (1.16) (2.23) (–0.34) (–0.05) (1.36) (1.51) Hostile resist (b$$_{\mathrm{2}})$$–0.065*** –0.050* –0.026* –0.032 (–2.68) (–1.85) (–1.67) (–1.26) Opposed hostile resist (b$$_{\mathrm{3}})$$0.077** 0.046** 0.032 0.038* (2.18) (2.41) (1.17) (1.84) Campaign characteristics Hedge fund offer dummy 0.005 0.010 –0.007 –0.003 0.012 0.014 –0.010 –0.008 (0.22) (0.48) (–0.38) (–0.20) (0.54) (0.63) (–0.43) (–0.34) Other takeover dummy –0.010 –0.008 –0.003 –0.001 0.002 0.003 –0.014 –0.013 (–1.09) (–0.93) (–0.24) (–0.10) (0.25) (0.30) (–1.43) (–1.32) Percentage owned –0.001 0.000 –0.004** –0.003** 0.001 0.001 –0.002* –0.002 (–0.56) (–0.15) (–2.26) (–2.07) (0.89) (1.04) (–1.74) (–1.57) Hedge fund in SEC group dummy 0.007 0.011 –0.037 –0.035 0.003 0.005 –0.044 –0.042 (0.36) (0.74) (–0.95) (–0.90) (0.15) (0.29) (–1.40) (–1.33) Other 13D filer dummy –0.014 –0.011 –0.005 –0.003 –0.014 –0.013 –0.000 0.002 (–1.00) (–0.79) (–0.68) (–0.38) (–1.53) (–1.38) (–0.02) (0.15) Other explicit support dummy 0.034 0.037 0.043** 0.044** 0.033** 0.034** 0.007 0.007 (1.53) (1.65) (2.31) (2.27) (2.33) (2.46) (0.32) (0.34) Hedge fund characteristics Order quartile –0.009 –0.010 0.004 0.003 –0.006 –0.007 0.001 –0.000 (–1.33) (–1.52) (0.71) (0.44) (–1.25) (–1.41) (0.18) (–0.02) HF prior counterresistance dummy 0.046*** 0.046*** 0.022 0.023 0.028** 0.028** 0.026 0.026 (3.93) (3.86) (1.04) (1.08) (2.38) (2.47) (1.52) (1.52) Firm characteristics Cash flow/assets or ROA/assets –0.469*** –0.468*** –0.701*** –0.698*** –0.453*** –0.455*** –0.490*** –0.492*** (–4.86) (–4.93) (–9.97) (–9.68) (–6.60) (–6.65) (–6.62) (–6.64) log of market cap. ( \$$MM) 0.009 0.010* 0.005 0.005 0.003 0.004 0.002 0.003 (1.61) (1.79) (1.05) (1.15) (0.85) (0.96) (0.65) (0.81) Cash –0.152*** –0.142*** –0.169*** –0.164*** –0.147*** –0.144*** –0.090*** –0.087*** (–4.35) (–3.48) (–3.17) (–3.00) (–7.26) (–6.65) (–3.22) (–3.02) Tobin’s q 0.003 0.002 0.004 0.003 0.001 0.000 0.005 0.005 (0.43) (0.31) (0.70) (0.63) (0.15) (0.05) (0.92) (0.80) RND –0.050 –0.055 –0.145 –0.145 –0.141 –0.145 –0.174* –0.175* (–0.29) (–0.33) (–1.15) (–1.17) (–1.32) (–1.37) (–1.76) (–1.79) Dividend yield –0.376 –0.401 –0.111 –0.114 –0.095 –0.106 –0.174 –0.187 (–1.41) (–1.48) (–0.54) (–0.61) (–0.64) (–0.67) (–1.09) (–1.09) Leverage –0.007 –0.010 –0.034 –0.036 0.005 0.003 –0.014 –0.016 (–0.37) (–0.63) (–1.20) (–1.26) (0.28) (0.19) (–0.71) (–0.79) HHI –0.048 –0.037 –0.036 –0.031 –0.066* –0.061* –0.048* –0.044* (–0.85) (–0.67) (–1.11) (–1.00) (–1.87) (–1.75) (–1.80) (–1.67) Liquidity 0.009 0.009 0.007 0.007 0.012 0.013 0.013* 0.013* (0.68) (0.75) (1.28) (1.66) (1.03) (1.07) (1.76) (1.93) Institutional ownership 0.032 0.024 0.024 0.019 0.038 0.035 0.030 0.027 (0.51) (0.39) (0.58) (0.45) (0.95) (0.84) (0.79) (0.68) Insider ownership –0.024 –0.027 –0.009 –0.011 –0.039 –0.040 –0.007 –0.007 (–0.52) (–0.56) (–0.19) (–0.22) (–1.11) (–1.12) (–0.22) (–0.22) Pill in force dummy –0.005 –0.010 –0.002 –0.006 0.002 –0.000 –0.000 –0.003 (–0.31) (–0.65) (–0.13) (–0.38) (0.35) (–0.02) (–0.02) (–0.34) b$$_{\mathrm{1 }}+$$ b$$_{\mathrm{2}}$$ –0.055** –0.028 –0.027 –0.014 (–2.03) (–1.55) (–1.59) (–0.76) b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}+$$ b$$_{\mathrm{3}}$$ 0.022 0.017 0.005 0.024 (0.99) (0.66) (0.26) (1.05) N 629 629 555 555 629 629 557 557 Includes year & industry dummies? Yes Yes Yes Yes Yes Yes Yes Yes Adjusted R$$^{\mathrm{2}}$$ 0.166 0.175 0.323 0.332 0.240 0.242 0.237 0.239 This table reports OLS regression results. The dependent variable is the change in cash flows or the change in ROA. The appendix defines the dependent and the control variables. Any resist is set to 1 for the 356 campaigns with moderate or hostile target resistance and 0 otherwise. Hostile resist is set to 1 for the subset of 207 hostile target resistance campaigns with or without formal hedge counterresistance and 0 otherwise. Opposed hostile resist is set to 1 for the subset of 138 hostile target resistance campaigns with formal hedge fund counterresistance and 0 otherwise. Regressions include year and 1-digit SIC industry dummies. Standard errors are clustered by time and hedge fund, and t-values are reported below the coefficients in parentheses. HHI, Herfindahl-Hirschman index***, **, and * denotes statistical significance for the 1%, 5%, and 10% levels, respectively. The coefficients on any resist in Columns 1 and 3 are insignificant, indicating no relation between resistance, either moderate or hostile, and a change in cash flow/assets. Columns 2 and 4, which include all resistance categorical variables, tell a different story. Campaigns with moderate resistance have 2-year changes in cash flows that are 2.2% higher than do campaigns with no resistance. Campaigns with unopposed hostile resistance have 1-year (2-year) performance that is 6.5% (5%) worse than campaigns with moderate resistance. Relative to campaigns with no resistance, these campaigns have a 1-year performance (b1$$+$$b2) that is 5.5% worse. The incremental effect of formal counterresistance is strictly positive: 7.7% (4.6%) for 1-year (2-year) changes. Relative to campaigns with no resistance, campaigns with opposed hostile resistance exhibit no significant differences in performance: the sum of all three resistance coefficients is positive, but insignificant. Results for changes in ROA in Columns 5 to 8 are slightly statistically weaker, but similar to results for changes in cash flows. Notably, changes in operating performance are positively related to both other explicit support and hedge fund (HF) prior counterresistance. For example, 2-year changes in cash flows are 4.4% higher with explicit support by other activists, consistent with the merger results in Table 5, and 1-year changes are 4.6% higher when the hedge fund has a history of formal counterresistance. These results imply that firms proactively make policy changes when under pressure, either by more than one activist or by an activist with a reputation for aggressive behavior. Untabulated tests find that the positive effect of formal hedge fund counterresistance occurs regardless of whether the counterresistance is definitively successful (definitive success involves the hedge fund winning the lawsuit, the unsolicited tender offer, or a proxy fight and occurs in about 30% of campaigns with formal hedge fund counterresistance). This result is consistent with that of Saffieddine and Titman (1999) and Boyson, Gantchev, and Shivdasani (2017), who examine attempted takeovers. The operating performance results differ somewhat from the merger results in that campaigns with moderate resistance have better operating performance than do campaigns with no resistance. One potential explanation is that moderate resistance leads to more active negotiations between target firms and hedge funds and more significant policy changes by targets in these campaigns. However, since these results only hold for changes in cash flows, we do not want to overemphasize them. 4. Rapid Exit and Resistance Since formal hedge fund counterresistance positively affects campaign outcomes, we further investigate why some hedge funds choose not to engage in formal counterresistance. One possible explanation is that because formal counterresistance is expensive, some hedge funds choose instead to exit rapidly and cut their losses. This idea relates to a large literature examining the relation between blockholder voice (activism) and blockholder exit (sale of shares). Additionally, we are interested in whether the negative resistance CARs are driven by shareholder expectations that hedge funds will quickly exit liquid stocks. It could be that activists only continue activism in less liquid stocks, for which exit is more likely to lead to a negative price impact, consistent with Bhide (1993) and Coffee (1991) and broadly consistent with the empirical findings of Bharath, Jayaraman, and Nagar (2013) and Edmans, Fang, and Zur (2013). To investigate these questions, we hand-collect exit dates for each campaign. We do not rely on Shark Repellant’s reported end dates, since their definition of exit typically indicates the resolution of the activist campaign and not the date the activist sells his shares. We review each campaign to determine the following potential means of exit: The hedge fund drops below 5% ownership. The hedge fund switches to filing a 13G (becomes a passive owner). The firm files for bankruptcy or involuntarily delists. The firm voluntarily delists (“goes dark”). The firm is acquired. The hedge fund is still active in the firm as of February 2017 (no exit). We create a dummy variable called early exit, which we set to 1 if the hedge fund exits the target firm within 1 year (or 6 months) of activism announcement, by method 1 or 2, without achieving the stated goals of activism or gaining a board seat and without the target firm merging before activist exit. Stated goals include requests for board representation, firm or asset sale, governance, management, or operating changes. We consider exit within 1 year to be early because the prior literature (e.g., Brav et al. 2008) finds that the average duration of an activist campaign is over 2 years. Our results do not depend on method 2 (filing a 13G), as these observations make up less than 2% of the total sample. Summary statistics regarding exit, reported in Table 7, panel A, indicate that in the full sample about 13% of hedge funds exit early within 1 year. Hedge funds in the unopposed hostile resist subsample exit early at a rate of 20%, a number significantly higher than the 4% early exit rate for the opposed hostile resist subsample. Table 7 Hostile resistance and the early exit decision A. Summary statistics With target resist All No target resist Moderate target resist Unopposed hostile resist Opposed hostile resist Differences (1) (2) (3) (4) (5) (2)-(3) (3)-(4) (4)-(5) Number of observations 821 465 149 69 138 NA NA NA Exit in 6 months, no success 0.074 0.084 0.060 0.116 0.036 0.024 –0.056 0.080* Exit in 1 year, no success 0.133 0.142 0.154 0.203 0.043 –0.013 –0.049 0.160*** Active in 2017 or active when firm leaves Compustat 0.364 0.361 0.342 0.362 0.399 0.019 –0.020 –0.037 Average days in activism, given voluntary exit 757 760 692 611 881 68 80 $$-270^{*}$$ A. Summary statistics With target resist All No target resist Moderate target resist Unopposed hostile resist Opposed hostile resist Differences (1) (2) (3) (4) (5) (2)-(3) (3)-(4) (4)-(5) Number of observations 821 465 149 69 138 NA NA NA Exit in 6 months, no success 0.074 0.084 0.060 0.116 0.036 0.024 –0.056 0.080* Exit in 1 year, no success 0.133 0.142 0.154 0.203 0.043 –0.013 –0.049 0.160*** Active in 2017 or active when firm leaves Compustat 0.364 0.361 0.342 0.362 0.399 0.019 –0.020 –0.037 Average days in activism, given voluntary exit 757 760 692 611 881 68 80 $$-270^{*}$$ B. Decision to exit early Exit in 6 months, no success Exit in 1 year, no success (1) (2) (3) (4) (5) (6) Resistance categorical variables Any resist (b$$_{\mathrm{1}})$$ –0.017 –0.018 –0.017 –0.007 0.018 0.017 (–1.11) (–0.80) (–0.77) (–0.38) (0.78) (0.75) Hostile resist (b$$_{\mathrm{2}})$$ 0.053 0.025 0.050 0.028 (1.47) (0.90) (1.03) (0.58) Opposed hostile resist (b$$_{\mathrm{3}})$$ –0.046* –0.033 –0.110*** –0.099** (–1.84) (–1.37) (–2.69) (–2.17) Campaign characteristics Hedge fund offer dummy 0.021 0.016 0.018 0.044 0.041 0.032 (0.61) (0.46) (0.58) (1.13) (0.91) (0.99) Other takeover dummy 0.026* 0.026** 0.025** 0.066*** 0.065*** 0.065*** (1.91) (2.01) (2.05) (3.30) (3.72) (3.70) Percentage owned –0.012 –0.014 –0.013 –0.023 –0.024 –0.024 (–0.97) (–1.18) (–1.34) (–1.39) (–1.43) (–1.57) Hedge fund in SEC group dummy –0.004 –0.003 –0.003 –0.019 –0.019 –0.020 (–0.14) (–0.09) (–0.15) (–0.35) (–0.38) (–0.41) Other 13D filer dummy 0.003 0.001 0.001 –0.009 –0.016 –0.014 (0.26) (0.10) (0.11) (–0.34) (–0.65) (–0.61) Other explicit support dummy –0.018 –0.017 –0.016 –0.054 –0.050 –0.050 (–1.52) (–1.44) (–1.49) (–1.37) (–1.24) (–1.30) Hedge fund characteristics Order quartile –0.084*** –0.079*** –0.071*** –0.023 –0.021 –0.020 (–4.15) (–4.06) (–4.45) (–0.74) (–0.68) (–0.63) HF prior counterresistance dummy 0.030 0.030 0.028 –0.007 –0.003 –0.002 (1.45) (1.52) (1.54) (–0.23) (–0.09) (–0.08) Firm characteristics Cash flow/assets –0.009** –0.007* –0.008** –0.019*** –0.017*** –0.017*** (–1.98) (–1.86) (–1.97) (–2.46) (–2.42) (–2.53) log of market cap. ($\$$MM) 0.007 0.006 0.002 –0.020 –0.021 –0.021 (0.60) (0.51) (0.25) (–0.96) (–0.99) (–0.95) Cash –0.017 –0.019 –0.019 –0.005 –0.009 –0.009 (–1.09) (–1.26) (–1.31) (–0.44) (–0.84) (–0.88) Tobin’s q 0.001 0.002 0.001 0.003 0.006** 0.006** (0.19) (0.41) (0.49) (0.96) (2.05) (2.16) RND 0.008 0.008 0.007 0.003 0.000 –0.000 (1.06) (1.24) (1.06) (0.23) (0.03) (–0.05) Dividend yield –0.002 –0.003 –0.000 –0.000 –0.000 –0.000 (–0.69) (–0.69) (–0.66) (–0.11) (–0.04) (–0.04) Leverage 0.008 0.008 0.008 0.014 0.013 0.013 (0.73) (0.76) (0.71) (0.89) (0.78) (0.79) HHI 0.001 0.001 0.003 0.019*** 0.016** 0.015** (0.18) (0.09) (0.04) (2.61) (2.24) (2.28) Liquidity 0.004 0.004 0.008 0.011** 0.010** 0.013** (1.13) (1.23) (1.29) (2.32) (2.20) (2.02) Liquidity x Hostile resist –0.003*** –0.003* (–2.58) (–1.86) Liquidity x Opposed hostile resist 0.001** 0.001** (2.12) (1.97) Institutional ownership –0.001 0.001 0.000 0.022 0.025 0.024 (–0.13) (0.05) (0.03) (1.08) (1.37) (1.36) Insider ownership 0.002 0.003 0.003 0.006 0.005 0.005 (1.02) (0.99) (1.09) (1.04) (0.86) (0.94) Pill in force dummy 0.004 0.009 0.009 0.017 0.023 0.023 (0.29) (0.66) (0.70) (0.58) (0.86) (0.86) b$$_{\mathrm{1}}+$$b$$_{\mathrm{2}}$$0.035 0.068* (1.04) (1.63) b$$_{\mathrm{1}}+$$b$$_{\mathrm{2}}+$$b$$_{\mathrm{3}}$$–0.011* –0.042** (–1.69) (–2.26) N 821 821 821 819 819 819 Includes year & industry dummies? Yes Yes Yes Yes Yes Yes Pseudo R$$^{\mathrm{2}}$$0.122 0.135 0.159 0.098 0.123 0.126 B. Decision to exit early Exit in 6 months, no success Exit in 1 year, no success (1) (2) (3) (4) (5) (6) Resistance categorical variables Any resist (b$$_{\mathrm{1}})$$–0.017 –0.018 –0.017 –0.007 0.018 0.017 (–1.11) (–0.80) (–0.77) (–0.38) (0.78) (0.75) Hostile resist (b$$_{\mathrm{2}})$$0.053 0.025 0.050 0.028 (1.47) (0.90) (1.03) (0.58) Opposed hostile resist (b$$_{\mathrm{3}})$$–0.046* –0.033 –0.110*** –0.099** (–1.84) (–1.37) (–2.69) (–2.17) Campaign characteristics Hedge fund offer dummy 0.021 0.016 0.018 0.044 0.041 0.032 (0.61) (0.46) (0.58) (1.13) (0.91) (0.99) Other takeover dummy 0.026* 0.026** 0.025** 0.066*** 0.065*** 0.065*** (1.91) (2.01) (2.05) (3.30) (3.72) (3.70) Percentage owned –0.012 –0.014 –0.013 –0.023 –0.024 –0.024 (–0.97) (–1.18) (–1.34) (–1.39) (–1.43) (–1.57) Hedge fund in SEC group dummy –0.004 –0.003 –0.003 –0.019 –0.019 –0.020 (–0.14) (–0.09) (–0.15) (–0.35) (–0.38) (–0.41) Other 13D filer dummy 0.003 0.001 0.001 –0.009 –0.016 –0.014 (0.26) (0.10) (0.11) (–0.34) (–0.65) (–0.61) Other explicit support dummy –0.018 –0.017 –0.016 –0.054 –0.050 –0.050 (–1.52) (–1.44) (–1.49) (–1.37) (–1.24) (–1.30) Hedge fund characteristics Order quartile –0.084*** –0.079*** –0.071*** –0.023 –0.021 –0.020 (–4.15) (–4.06) (–4.45) (–0.74) (–0.68) (–0.63) HF prior counterresistance dummy 0.030 0.030 0.028 –0.007 –0.003 –0.002 (1.45) (1.52) (1.54) (–0.23) (–0.09) (–0.08) Firm characteristics Cash flow/assets –0.009** –0.007* –0.008** –0.019*** –0.017*** –0.017*** (–1.98) (–1.86) (–1.97) (–2.46) (–2.42) (–2.53) log of market cap. ( \$$MM) 0.007 0.006 0.002 –0.020 –0.021 –0.021 (0.60) (0.51) (0.25) (–0.96) (–0.99) (–0.95) Cash –0.017 –0.019 –0.019 –0.005 –0.009 –0.009 (–1.09) (–1.26) (–1.31) (–0.44) (–0.84) (–0.88) Tobin’s q 0.001 0.002 0.001 0.003 0.006** 0.006** (0.19) (0.41) (0.49) (0.96) (2.05) (2.16) RND 0.008 0.008 0.007 0.003 0.000 –0.000 (1.06) (1.24) (1.06) (0.23) (0.03) (–0.05) Dividend yield –0.002 –0.003 –0.000 –0.000 –0.000 –0.000 (–0.69) (–0.69) (–0.66) (–0.11) (–0.04) (–0.04) Leverage 0.008 0.008 0.008 0.014 0.013 0.013 (0.73) (0.76) (0.71) (0.89) (0.78) (0.79) HHI 0.001 0.001 0.003 0.019*** 0.016** 0.015** (0.18) (0.09) (0.04) (2.61) (2.24) (2.28) Liquidity 0.004 0.004 0.008 0.011** 0.010** 0.013** (1.13) (1.23) (1.29) (2.32) (2.20) (2.02) Liquidity x Hostile resist –0.003*** –0.003* (–2.58) (–1.86) Liquidity x Opposed hostile resist 0.001** 0.001** (2.12) (1.97) Institutional ownership –0.001 0.001 0.000 0.022 0.025 0.024 (–0.13) (0.05) (0.03) (1.08) (1.37) (1.36) Insider ownership 0.002 0.003 0.003 0.006 0.005 0.005 (1.02) (0.99) (1.09) (1.04) (0.86) (0.94) Pill in force dummy 0.004 0.009 0.009 0.017 0.023 0.023 (0.29) (0.66) (0.70) (0.58) (0.86) (0.86) b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}$$ 0.035 0.068* (1.04) (1.63) b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}+$$ b$$_{\mathrm{3}}$$ –0.011* –0.042** (–1.69) (–2.26) N 821 821 821 819 819 819 Includes year & industry dummies? Yes Yes Yes Yes Yes Yes Pseudo R$$^{\mathrm{2}}$$ 0.122 0.135 0.159 0.098 0.123 0.126 This table summarizes the decisions of hedge fund activists to exit activism campaigns. “No success” means that the activist did not achieve any stated activism goals or obtain board representation and the target firm did not enter into a merger agreement or was not acquired before the hedge fund exited the campaign. Panel A reports summary statistics. Exit in 6 months, no success (Exit in 1 year, no success) is a dummy variable set to 1 if the hedge fund achieves no success and drops below 5% ownership or switches to filing a 13G form within 6 months (1 year) of the activism start date, and 0 otherwise. Active in 2017 or active when firm leaves Compustat is a dummy variable set to 1 if the hedge fund is still active in the target firm as of February 2017 or is still active when the firm leaves Compustat through a merger, delisting, liquidation, or bankruptcy. Average days in activism, given voluntary exit is the number of days between the announcement of activism and the time that the hedge fund ceases activism by dropping below 5% ownership or filing a 13G form, reported only for the subsample of campaigns in which the hedge fund exits the target firm. Activism campaigns are divided into four mutually exclusive categories: Moderate target resist includes the 149 campaigns in which the target firm resists in a moderate manner; Unopposed hostile resist includes 69 campaigns in which the target firm resists in a hostile manner and the hedge fund does not counterresist; Opposed hostile resist includes 138 campaigns in which the target firm resists in a hostile manner and the hedge fund counterresists; and No target resist includes 465 campaigns in which resistance is not observed. Panel B reports the results of logit analyses modeling the exit decision. The values reported in panel B represent the probability change for a move from 0 to 1 for indicator variables and from the 25th to the 75th percentile for continuous variables, setting all other variables at their means. The dependent variable in Columns 1 to 3 is Exit in 6 months, no success, and the dependent variable in Columns 4 to 6 is Exit in 1 year, no success. Any resist is set to 1 for the 356 campaigns with moderate or hostile target resistance and 0 otherwise. Hostile resist is set to 1 for the subset of 207 resistance campaigns with hostile target resistance and 0 otherwise. Opposed hostile resist is set to 1 for the subset of 138 resistance campaigns with hostile target resistance and hedge fund counterresistance and 0 otherwise. The appendix describes all other control variables. All regressions include year dummies and 1-digit SIC industry dummies. Standard errors are clustered by time and hedge fund, and z-values are reported below the coefficients in parentheses. NA, not applicable; HHI, Herfindahl-Hirschman index. ***, **, and * denotes statistical significance for the 1%, 5%, and 10% levels, respectively. Table 7 Hostile resistance and the early exit decision A. Summary statistics With target resist All No target resist Moderate target resist Unopposed hostile resist Opposed hostile resist Differences (1) (2) (3) (4) (5) (2)-(3) (3)-(4) (4)-(5) Number of observations 821 465 149 69 138 NA NA NA Exit in 6 months, no success 0.074 0.084 0.060 0.116 0.036 0.024 –0.056 0.080* Exit in 1 year, no success 0.133 0.142 0.154 0.203 0.043 –0.013 –0.049 0.160*** Active in 2017 or active when firm leaves Compustat 0.364 0.361 0.342 0.362 0.399 0.019 –0.020 –0.037 Average days in activism, given voluntary exit 757 760 692 611 881 68 80 $$-270^{*}$$ A. Summary statistics With target resist All No target resist Moderate target resist Unopposed hostile resist Opposed hostile resist Differences (1) (2) (3) (4) (5) (2)-(3) (3)-(4) (4)-(5) Number of observations 821 465 149 69 138 NA NA NA Exit in 6 months, no success 0.074 0.084 0.060 0.116 0.036 0.024 –0.056 0.080* Exit in 1 year, no success 0.133 0.142 0.154 0.203 0.043 –0.013 –0.049 0.160*** Active in 2017 or active when firm leaves Compustat 0.364 0.361 0.342 0.362 0.399 0.019 –0.020 –0.037 Average days in activism, given voluntary exit 757 760 692 611 881 68 80 $$-270^{*}$$ B. Decision to exit early Exit in 6 months, no success Exit in 1 year, no success (1) (2) (3) (4) (5) (6) Resistance categorical variables Any resist (b$$_{\mathrm{1}})$$ –0.017 –0.018 –0.017 –0.007 0.018 0.017 (–1.11) (–0.80) (–0.77) (–0.38) (0.78) (0.75) Hostile resist (b$$_{\mathrm{2}})$$ 0.053 0.025 0.050 0.028 (1.47) (0.90) (1.03) (0.58) Opposed hostile resist (b$$_{\mathrm{3}})$$ –0.046* –0.033 –0.110*** –0.099** (–1.84) (–1.37) (–2.69) (–2.17) Campaign characteristics Hedge fund offer dummy 0.021 0.016 0.018 0.044 0.041 0.032 (0.61) (0.46) (0.58) (1.13) (0.91) (0.99) Other takeover dummy 0.026* 0.026** 0.025** 0.066*** 0.065*** 0.065*** (1.91) (2.01) (2.05) (3.30) (3.72) (3.70) Percentage owned –0.012 –0.014 –0.013 –0.023 –0.024 –0.024 (–0.97) (–1.18) (–1.34) (–1.39) (–1.43) (–1.57) Hedge fund in SEC group dummy –0.004 –0.003 –0.003 –0.019 –0.019 –0.020 (–0.14) (–0.09) (–0.15) (–0.35) (–0.38) (–0.41) Other 13D filer dummy 0.003 0.001 0.001 –0.009 –0.016 –0.014 (0.26) (0.10) (0.11) (–0.34) (–0.65) (–0.61) Other explicit support dummy –0.018 –0.017 –0.016 –0.054 –0.050 –0.050 (–1.52) (–1.44) (–1.49) (–1.37) (–1.24) (–1.30) Hedge fund characteristics Order quartile –0.084*** –0.079*** –0.071*** –0.023 –0.021 –0.020 (–4.15) (–4.06) (–4.45) (–0.74) (–0.68) (–0.63) HF prior counterresistance dummy 0.030 0.030 0.028 –0.007 –0.003 –0.002 (1.45) (1.52) (1.54) (–0.23) (–0.09) (–0.08) Firm characteristics Cash flow/assets –0.009** –0.007* –0.008** –0.019*** –0.017*** –0.017*** (–1.98) (–1.86) (–1.97) (–2.46) (–2.42) (–2.53) log of market cap. ( $\$$MM) 0.007 0.006 0.002 –0.020 –0.021 –0.021 (0.60) (0.51) (0.25) (–0.96) (–0.99) (–0.95) Cash –0.017 –0.019 –0.019 –0.005 –0.009 –0.009 (–1.09) (–1.26) (–1.31) (–0.44) (–0.84) (–0.88) Tobin’s q 0.001 0.002 0.001 0.003 0.006** 0.006** (0.19) (0.41) (0.49) (0.96) (2.05) (2.16) RND 0.008 0.008 0.007 0.003 0.000 –0.000 (1.06) (1.24) (1.06) (0.23) (0.03) (–0.05) Dividend yield –0.002 –0.003 –0.000 –0.000 –0.000 –0.000 (–0.69) (–0.69) (–0.66) (–0.11) (–0.04) (–0.04) Leverage 0.008 0.008 0.008 0.014 0.013 0.013 (0.73) (0.76) (0.71) (0.89) (0.78) (0.79) HHI 0.001 0.001 0.003 0.019*** 0.016** 0.015** (0.18) (0.09) (0.04) (2.61) (2.24) (2.28) Liquidity 0.004 0.004 0.008 0.011** 0.010** 0.013** (1.13) (1.23) (1.29) (2.32) (2.20) (2.02) Liquidity x Hostile resist –0.003*** –0.003* (–2.58) (–1.86) Liquidity x Opposed hostile resist 0.001** 0.001** (2.12) (1.97) Institutional ownership –0.001 0.001 0.000 0.022 0.025 0.024 (–0.13) (0.05) (0.03) (1.08) (1.37) (1.36) Insider ownership 0.002 0.003 0.003 0.006 0.005 0.005 (1.02) (0.99) (1.09) (1.04) (0.86) (0.94) Pill in force dummy 0.004 0.009 0.009 0.017 0.023 0.023 (0.29) (0.66) (0.70) (0.58) (0.86) (0.86) b$$_{\mathrm{1}}+$$b$$_{\mathrm{2}}$$0.035 0.068* (1.04) (1.63) b$$_{\mathrm{1}}+$$b$$_{\mathrm{2}}+$$b$$_{\mathrm{3}}$$–0.011* –0.042** (–1.69) (–2.26) N 821 821 821 819 819 819 Includes year & industry dummies? Yes Yes Yes Yes Yes Yes Pseudo R$$^{\mathrm{2}}$$0.122 0.135 0.159 0.098 0.123 0.126 B. Decision to exit early Exit in 6 months, no success Exit in 1 year, no success (1) (2) (3) (4) (5) (6) Resistance categorical variables Any resist (b$$_{\mathrm{1}})$$–0.017 –0.018 –0.017 –0.007 0.018 0.017 (–1.11) (–0.80) (–0.77) (–0.38) (0.78) (0.75) Hostile resist (b$$_{\mathrm{2}})$$0.053 0.025 0.050 0.028 (1.47) (0.90) (1.03) (0.58) Opposed hostile resist (b$$_{\mathrm{3}})$$–0.046* –0.033 –0.110*** –0.099** (–1.84) (–1.37) (–2.69) (–2.17) Campaign characteristics Hedge fund offer dummy 0.021 0.016 0.018 0.044 0.041 0.032 (0.61) (0.46) (0.58) (1.13) (0.91) (0.99) Other takeover dummy 0.026* 0.026** 0.025** 0.066*** 0.065*** 0.065*** (1.91) (2.01) (2.05) (3.30) (3.72) (3.70) Percentage owned –0.012 –0.014 –0.013 –0.023 –0.024 –0.024 (–0.97) (–1.18) (–1.34) (–1.39) (–1.43) (–1.57) Hedge fund in SEC group dummy –0.004 –0.003 –0.003 –0.019 –0.019 –0.020 (–0.14) (–0.09) (–0.15) (–0.35) (–0.38) (–0.41) Other 13D filer dummy 0.003 0.001 0.001 –0.009 –0.016 –0.014 (0.26) (0.10) (0.11) (–0.34) (–0.65) (–0.61) Other explicit support dummy –0.018 –0.017 –0.016 –0.054 –0.050 –0.050 (–1.52) (–1.44) (–1.49) (–1.37) (–1.24) (–1.30) Hedge fund characteristics Order quartile –0.084*** –0.079*** –0.071*** –0.023 –0.021 –0.020 (–4.15) (–4.06) (–4.45) (–0.74) (–0.68) (–0.63) HF prior counterresistance dummy 0.030 0.030 0.028 –0.007 –0.003 –0.002 (1.45) (1.52) (1.54) (–0.23) (–0.09) (–0.08) Firm characteristics Cash flow/assets –0.009** –0.007* –0.008** –0.019*** –0.017*** –0.017*** (–1.98) (–1.86) (–1.97) (–2.46) (–2.42) (–2.53) log of market cap. ( \$$MM) 0.007 0.006 0.002 –0.020 –0.021 –0.021 (0.60) (0.51) (0.25) (–0.96) (–0.99) (–0.95) Cash –0.017 –0.019 –0.019 –0.005 –0.009 –0.009 (–1.09) (–1.26) (–1.31) (–0.44) (–0.84) (–0.88) Tobin’s q 0.001 0.002 0.001 0.003 0.006** 0.006** (0.19) (0.41) (0.49) (0.96) (2.05) (2.16) RND 0.008 0.008 0.007 0.003 0.000 –0.000 (1.06) (1.24) (1.06) (0.23) (0.03) (–0.05) Dividend yield –0.002 –0.003 –0.000 –0.000 –0.000 –0.000 (–0.69) (–0.69) (–0.66) (–0.11) (–0.04) (–0.04) Leverage 0.008 0.008 0.008 0.014 0.013 0.013 (0.73) (0.76) (0.71) (0.89) (0.78) (0.79) HHI 0.001 0.001 0.003 0.019*** 0.016** 0.015** (0.18) (0.09) (0.04) (2.61) (2.24) (2.28) Liquidity 0.004 0.004 0.008 0.011** 0.010** 0.013** (1.13) (1.23) (1.29) (2.32) (2.20) (2.02) Liquidity x Hostile resist –0.003*** –0.003* (–2.58) (–1.86) Liquidity x Opposed hostile resist 0.001** 0.001** (2.12) (1.97) Institutional ownership –0.001 0.001 0.000 0.022 0.025 0.024 (–0.13) (0.05) (0.03) (1.08) (1.37) (1.36) Insider ownership 0.002 0.003 0.003 0.006 0.005 0.005 (1.02) (0.99) (1.09) (1.04) (0.86) (0.94) Pill in force dummy 0.004 0.009 0.009 0.017 0.023 0.023 (0.29) (0.66) (0.70) (0.58) (0.86) (0.86) b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}$$ 0.035 0.068* (1.04) (1.63) b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}+$$ b$$_{\mathrm{3}}$$ –0.011* –0.042** (–1.69) (–2.26) N 821 821 821 819 819 819 Includes year & industry dummies? Yes Yes Yes Yes Yes Yes Pseudo R$$^{\mathrm{2}}$$ 0.122 0.135 0.159 0.098 0.123 0.126 This table summarizes the decisions of hedge fund activists to exit activism campaigns. “No success” means that the activist did not achieve any stated activism goals or obtain board representation and the target firm did not enter into a merger agreement or was not acquired before the hedge fund exited the campaign. Panel A reports summary statistics. Exit in 6 months, no success (Exit in 1 year, no success) is a dummy variable set to 1 if the hedge fund achieves no success and drops below 5% ownership or switches to filing a 13G form within 6 months (1 year) of the activism start date, and 0 otherwise. Active in 2017 or active when firm leaves Compustat is a dummy variable set to 1 if the hedge fund is still active in the target firm as of February 2017 or is still active when the firm leaves Compustat through a merger, delisting, liquidation, or bankruptcy. Average days in activism, given voluntary exit is the number of days between the announcement of activism and the time that the hedge fund ceases activism by dropping below 5% ownership or filing a 13G form, reported only for the subsample of campaigns in which the hedge fund exits the target firm. Activism campaigns are divided into four mutually exclusive categories: Moderate target resist includes the 149 campaigns in which the target firm resists in a moderate manner; Unopposed hostile resist includes 69 campaigns in which the target firm resists in a hostile manner and the hedge fund does not counterresist; Opposed hostile resist includes 138 campaigns in which the target firm resists in a hostile manner and the hedge fund counterresists; and No target resist includes 465 campaigns in which resistance is not observed. Panel B reports the results of logit analyses modeling the exit decision. The values reported in panel B represent the probability change for a move from 0 to 1 for indicator variables and from the 25th to the 75th percentile for continuous variables, setting all other variables at their means. The dependent variable in Columns 1 to 3 is Exit in 6 months, no success, and the dependent variable in Columns 4 to 6 is Exit in 1 year, no success. Any resist is set to 1 for the 356 campaigns with moderate or hostile target resistance and 0 otherwise. Hostile resist is set to 1 for the subset of 207 resistance campaigns with hostile target resistance and 0 otherwise. Opposed hostile resist is set to 1 for the subset of 138 resistance campaigns with hostile target resistance and hedge fund counterresistance and 0 otherwise. The appendix describes all other control variables. All regressions include year dummies and 1-digit SIC industry dummies. Standard errors are clustered by time and hedge fund, and z-values are reported below the coefficients in parentheses. NA, not applicable; HHI, Herfindahl-Hirschman index. ***, **, and * denotes statistical significance for the 1%, 5%, and 10% levels, respectively. Rates of early exit for unopposed hostile resistance are higher than for no resistance and moderate resistance, but these differences are insignificant. Although early exit rates are higher for unopposed hostile resistance, the aggregate early exit rates are still low. In 36% of campaigns, there is no voluntary exit: hedge funds are still involved in target firms as of February 2017 (8%) or are involved in the target firm when it is acquired (25%) or delists (3%).26 Given voluntary exit, hedge funds stay active in target firms for 757 days, or just over 2 years. The mean time in activism is significantly lower, by 9 months, for the unopposed hostile resist subsample relative to the opposed hostile resist subsample. Panel B presents logit regressions in which the dependent variable is early exit. Regressions include all campaigns (Columns 4–6 have 819 observations because the dependent variable is perfectly correlated with a year dummy). Instead of regression coefficients, the table reports the raw probability change for a move from 0 to 1 for indicator variables and from the 25th percentile to the 75th percentile for continuous variables, with all other variables at their means. Regressions include year and one-digit SIC industry dummies; standard errors are clustered by year and hedge fund; and z-values are reported in parentheses. As a benchmark, the unconditional probability of earlyexit in 6 months is 7%, and the unconditional probability of earlyexit in 1 year is 13%. Because results are similar for 6-month and 1-year early exit, we focus on the 1-year results. Results in Column 4 indicate that campaigns with any resistance are no more likely to experience early exit than campaigns with no resistance. Control variables indicate that takeover-related campaigns are more likely to experience early exit, as are targets with more liquid stocks. Column 5 adds the other categorical resistance variables and shows that campaigns with moderate resistance are no more likely to experience early exit than are campaigns with no resistance; campaigns with unopposed hostile resistance are no more likely to experience early exit than are campaigns with moderate resistance; and campaigns with opposed hostile resistance are a relative 85% less likely (0.11/0.13) to experience early exit relative to campaigns with unopposed hostile resistance. The sums of the coefficients indicate that relative to campaigns with no resistance, the probability of early exit is a relative 52% higher (0.068/0.13) for campaigns with unopposed hostile resistance and a relative 32% lower (0.042/0.13) for campaigns with opposed hostile resistance. Hence, hedge funds that do not choose to formally oppose hostile target resistance are more likely to exit their campaigns early. Finally, in an untabulated robustness check, we find that our results indicating worse operating performance and a lower likelihood of merger for the unopposed hostile resist campaigns are not driven by campaigns in which hedge funds exit early. Next, we address the question of whether the average negative resistance CAR for the unopposed hostile resist subsample is driven by shareholder expectations that hedge funds will quickly exit the most liquid stocks. Results in Column 5 indicate that a change in liquidity from the 25th percentile to 75th percentile results in an 8% relative increase (0.01/0.13) in the probability of early exit. This result for the full sample is consistent with the prior literature. To test whether this result holds for the unopposed hostile resist subsample, Column 6 includes interactions between liquidity and the two hostile resist dummies. If hedge funds in the unopposed hostile resist subsample are more likely to quickly exit liquid stocks, we should observe a positive and significant coefficient on the interaction of liquidity and the hostile resist dummy. However, we find the opposite result: the negative and significant coefficient on the interaction between liquidity and hostile resist indicates that hedge funds that do not formally counterresist are more likely to exit early from less liquid stocks. The results for this subsample are instead consistent with the idea that the negative market reaction to hostile resistance is driven by investor expectations that hostile resistance will reduce the expected benefits to activism. While our exit and liquidity results are inconsistent with the prior literature, we note that the past literature does not segregate cases in which target firms directly oppose blockholders. Further, much of the prior literature focuses on voice and exit among both active and passive blockholders, and not on the unique role played by hedge fund activists. Hence, while our interpretation of this result is consistent with the data we present, the relation between hedge fund activism and target firm liquidity remains a topic for future research. 5. Long-term Stock Performance and Resistance Next, we examine estimated activist holding-period returns. We perform this analysis for several reasons. First, we have shown that short-term market returns are positive for the activism announcement, negative for the hostile target firm resistance announcement, and positive for the hedge fund opposition announcement. However, we have not formally investigated the long-term stock market response to resistance and counterresistance. We are primarily interested in whether hostile target resistance reduces the positive expected return to activism and whether hedge fund opposition restores it. Second, we have shown that activists that do not formally oppose hostile target resistance are more likely to exit early, hypothesizing that these funds rationally cut their losses to avoid investing additional resources in activism. Estimating the total returns to activists for this subset can help quantify this finding. Finally, the prior literature (e.g., Brav et al. 2008; Klein and Zur 2009; Bebchuk, Brav, and Jiang 2015; Boyson, Gantchev, and Shivdasani 2017) finds that the initial positive stock market reaction to activism tends to persist over the activist holding period, and we wish to test whether this result varies by resistance category. Univariate statistics in Table 1 indicate that BHARs are on average positive and significantly greater than zero for all subsamples except the unopposed hostile resist subsample. Results are directionally similar, although magnitudes are higher, for holding-period CARs. BHARs are highest (and quite similar in magnitude) for opposed hostile target resistance and for moderate target resistance. As discussed at the end of Section 1, the small net positive announcement effect of activism, hostile target resistance, and formal hedge fund counterresistance is not predictive of the high activist holding-period positive returns for opposed hostile resistance. Similarly, the net positive announcement effect of activism and hostile target resistance for the unopposed hostile resistance subsample does not fully predict the sharply lower holding-period returns for this subsample. Table 8 presents multivariate analyses of activist holding-period returns. In Columns 1 and 2, the dependent variable is the BHAR, and, in Columns 3 and 4, the dependent variable is the holding-period CAR measured in excess of the value-weighted CRSP index. These returns are calculated using daily data measured from the day before activism is announced to the day after hedge fund exit (when we can observe the exit), or through the end of February 2017 for campaigns that are still active at this time.27 Regressions include year and one-digit SIC industry dummies; standard errors are clustered by year and hedge fund; and t-values are reported in parentheses. Table 8 Resistance and long-term stock market returns to activism Buy and hold returns$$_{\mathrm{{[-1\ {\rm act}, +1\ {\rm exit}]}}}$$ Holding-period CAR in excess of value-weighted CRSP index$$_{\mathrm{{[-1\ {\rm act}, +1\ {\rm exit}]}}}$$ (1) (2) (3) (4) Resistance categorical variables Any resist (b$$_{\mathrm{1}})$$ 0.005 0.049* 0.042 0.029 (0.10) (1.65) (0.67) (0.36) Hostile resist (b$$_{\mathrm{2}})$$ –0.222*** –0.086 (–2.86) (–0.99) Opposed hostile resist (b$$_{\mathrm{3}})$$ 0.205** 0.156*** (2.08) (2.50) Campaign characteristics Hedge fund offer dummy 0.277** 0.285** 0.175** 0.181** (1.98) (2.04) (1.95) (2.13) Other takeover dummy 0.089 0.087 0.053 0.049 (1.18) (1.16) (1.00) (0.89) Percentage owned –0.011** –0.010* –0.006 –0.006 (–2.19) (–1.89) (–1.24) (–1.10) Hedge fund in SEC group dummy 0.166 0.174 0.093* 0.103** (1.23) (1.27) (1.82) (2.11) Other 13D filer dummy –0.091 –0.088 –0.090*** –0.084*** (–1.50) (–1.42) (–3.16) (–2.74) Other explicit support dummy 0.008 0.012 0.004 0.004 (0.14) (0.21) (0.08) (0.09) Hedge fund characteristics Order quartile –0.009 –0.013 0.015 0.015 (–0.33) (–0.46) (0.43) (0.43) HF prior counterresistance dummy –0.010 –0.012 –0.034 –0.039 (–0.16) (–0.18) (–0.38) (–0.42) Firm characteristics Cash flow/assets –0.180 –0.165 –0.073 –0.065 (–0.64) (–0.62) (–0.28) (–0.26) log of market cap. ($\$$MM) –0.009 –0.006 –0.049*** –0.049** (–0.87) (–0.56) (–2.48) (–2.27) Cash –0.202 –0.168 –0.304* –0.290* (–0.94) (–0.79) (–1.85) (–1.80) Tobin’s q 0.021 0.022 0.027 0.026 (1.15) (1.15) (1.28) (1.25) RND 0.522 0.485 0.916*** 0.916*** (1.06) (1.03) (2.71) (2.84) Dividend yield –0.465 –0.538 –0.930 –1.029 (–0.27) (–0.33) (–0.94) (–1.13) Leverage –0.134* –0.137* –0.063 –0.064 (–1.87) (–1.94) (–0.68) (–0.69) HHI –0.181 –0.153 –0.084 –0.064 (–0.75) (–0.61) (–0.46) (–0.34) Liquidity 0.057 0.055* 0.000 0.000 (1.59) (1.64) (0.01) (0.01) Institutional ownership –0.200* –0.220* –0.228*** –0.235*** (–1.63) (–1.82) (–2.69) (–3.09) Insider ownership –0.256 –0.272* –0.043 –0.038 (–1.58) (–1.77) (–0.59) (–0.85) Pill in force dummy 0.051 0.035 0.023 0.016 (0.88) (0.59) (0.34) (0.25) b$$_{\mathrm{1}}+$$b$$_{\mathrm{2}}$$–0.173* –0.056 (–1.84) (–0.63) b$$_{\mathrm{1}}+$$b$$_{\mathrm{2}}+$$b$$_{\mathrm{3}}$$0.031 0.100*** (0.45) (1.80) N 726 726 726 726 Includes year & industry dummies? Yes Yes Yes Yes Adjusted R$$^{\mathrm{2}}$$0.106 0.109 0.111 0.111 Buy and hold returns$$_{\mathrm{{[-1\ {\rm act}, +1\ {\rm exit}]}}}$$Holding-period CAR in excess of value-weighted CRSP index$$_{\mathrm{{[-1\ {\rm act}, +1\ {\rm exit}]}}}$$(1) (2) (3) (4) Resistance categorical variables Any resist (b$$_{\mathrm{1}})$$0.005 0.049* 0.042 0.029 (0.10) (1.65) (0.67) (0.36) Hostile resist (b$$_{\mathrm{2}})$$–0.222*** –0.086 (–2.86) (–0.99) Opposed hostile resist (b$$_{\mathrm{3}})$$0.205** 0.156*** (2.08) (2.50) Campaign characteristics Hedge fund offer dummy 0.277** 0.285** 0.175** 0.181** (1.98) (2.04) (1.95) (2.13) Other takeover dummy 0.089 0.087 0.053 0.049 (1.18) (1.16) (1.00) (0.89) Percentage owned –0.011** –0.010* –0.006 –0.006 (–2.19) (–1.89) (–1.24) (–1.10) Hedge fund in SEC group dummy 0.166 0.174 0.093* 0.103** (1.23) (1.27) (1.82) (2.11) Other 13D filer dummy –0.091 –0.088 –0.090*** –0.084*** (–1.50) (–1.42) (–3.16) (–2.74) Other explicit support dummy 0.008 0.012 0.004 0.004 (0.14) (0.21) (0.08) (0.09) Hedge fund characteristics Order quartile –0.009 –0.013 0.015 0.015 (–0.33) (–0.46) (0.43) (0.43) HF prior counterresistance dummy –0.010 –0.012 –0.034 –0.039 (–0.16) (–0.18) (–0.38) (–0.42) Firm characteristics Cash flow/assets –0.180 –0.165 –0.073 –0.065 (–0.64) (–0.62) (–0.28) (–0.26) log of market cap. ( \$$MM) –0.009 –0.006 –0.049*** –0.049** (–0.87) (–0.56) (–2.48) (–2.27) Cash –0.202 –0.168 –0.304* –0.290* (–0.94) (–0.79) (–1.85) (–1.80) Tobin’s q 0.021 0.022 0.027 0.026 (1.15) (1.15) (1.28) (1.25) RND 0.522 0.485 0.916*** 0.916*** (1.06) (1.03) (2.71) (2.84) Dividend yield –0.465 –0.538 –0.930 –1.029 (–0.27) (–0.33) (–0.94) (–1.13) Leverage –0.134* –0.137* –0.063 –0.064 (–1.87) (–1.94) (–0.68) (–0.69) HHI –0.181 –0.153 –0.084 –0.064 (–0.75) (–0.61) (–0.46) (–0.34) Liquidity 0.057 0.055* 0.000 0.000 (1.59) (1.64) (0.01) (0.01) Institutional ownership –0.200* –0.220* –0.228*** –0.235*** (–1.63) (–1.82) (–2.69) (–3.09) Insider ownership –0.256 –0.272* –0.043 –0.038 (–1.58) (–1.77) (–0.59) (–0.85) Pill in force dummy 0.051 0.035 0.023 0.016 (0.88) (0.59) (0.34) (0.25) b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}$$ –0.173* –0.056 (–1.84) (–0.63) b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}+$$ b$$_{\mathrm{3}}$$ 0.031 0.100*** (0.45) (1.80) N 726 726 726 726 Includes year & industry dummies? Yes Yes Yes Yes Adjusted R$$^{\mathrm{2}}$$ 0.106 0.109 0.111 0.111 This table reports OLS regressions in which the dependent variable is either buy and hold returns (BHARs) (Columns 1 and 2) or holding-period CARs, both measured in excess of the value-weighted CRSP index$$_{\mathrm{[-1\ {\rm act},+1\ {\rm exit}]}}$$ (Columns 3 and 4). Means of exit include dropping below 5%; filing a 13G form; or being active in the firm when it delists, liquidates or declares bankruptcy, or merges. Regressions also include campaigns in which the hedge fund is still active in the target firm as of February 2017. For these latter campaigns, we calculate returns through February 28, 2017. All returns use daily data and include the period starting 1 day before the activism announcement and ending 1 day after the hedge fund exit. Any resist is set to 1 for the 356 campaigns with moderate or hostile target resistance and 0 otherwise. Hostile resist is set to 1 one for the subset of 207 resistance campaigns with hostile target resistance and 0 otherwise. Opposed hostile resist is set to 1 one for the subset of 138 campaigns with hostile target resistance and hedge fund counterresistance and 0 otherwise. The appendix describes all other control variables. Regressions include year and 1-digit SIC industry dummies. Standard errors are clustered by time and hedge fund, and t-values are reported below the coefficients in parentheses. HHI, Herfindahl-Hirschman index. ***, **, and * denotes statistical significance for the 1%, 5%, and 10% levels, respectively. Table 8 Resistance and long-term stock market returns to activism Buy and hold returns$$_{\mathrm{{[-1\ {\rm act}, +1\ {\rm exit}]}}}$$ Holding-period CAR in excess of value-weighted CRSP index$$_{\mathrm{{[-1\ {\rm act}, +1\ {\rm exit}]}}}$$ (1) (2) (3) (4) Resistance categorical variables Any resist (b$$_{\mathrm{1}})$$ 0.005 0.049* 0.042 0.029 (0.10) (1.65) (0.67) (0.36) Hostile resist (b$$_{\mathrm{2}})$$ –0.222*** –0.086 (–2.86) (–0.99) Opposed hostile resist (b$$_{\mathrm{3}})$$ 0.205** 0.156*** (2.08) (2.50) Campaign characteristics Hedge fund offer dummy 0.277** 0.285** 0.175** 0.181** (1.98) (2.04) (1.95) (2.13) Other takeover dummy 0.089 0.087 0.053 0.049 (1.18) (1.16) (1.00) (0.89) Percentage owned –0.011** –0.010* –0.006 –0.006 (–2.19) (–1.89) (–1.24) (–1.10) Hedge fund in SEC group dummy 0.166 0.174 0.093* 0.103** (1.23) (1.27) (1.82) (2.11) Other 13D filer dummy –0.091 –0.088 –0.090*** –0.084*** (–1.50) (–1.42) (–3.16) (–2.74) Other explicit support dummy 0.008 0.012 0.004 0.004 (0.14) (0.21) (0.08) (0.09) Hedge fund characteristics Order quartile –0.009 –0.013 0.015 0.015 (–0.33) (–0.46) (0.43) (0.43) HF prior counterresistance dummy –0.010 –0.012 –0.034 –0.039 (–0.16) (–0.18) (–0.38) (–0.42) Firm characteristics Cash flow/assets –0.180 –0.165 –0.073 –0.065 (–0.64) (–0.62) (–0.28) (–0.26) log of market cap. ( \$\$$MM) –0.009 –0.006 –0.049*** –0.049** (–0.87) (–0.56) (–2.48) (–2.27) Cash –0.202 –0.168 –0.304* –0.290* (–0.94) (–0.79) (–1.85) (–1.80) Tobin’s q 0.021 0.022 0.027 0.026 (1.15) (1.15) (1.28) (1.25) RND 0.522 0.485 0.916*** 0.916*** (1.06) (1.03) (2.71) (2.84) Dividend yield –0.465 –0.538 –0.930 –1.029 (–0.27) (–0.33) (–0.94) (–1.13) Leverage –0.134* –0.137* –0.063 –0.064 (–1.87) (–1.94) (–0.68) (–0.69) HHI –0.181 –0.153 –0.084 –0.064 (–0.75) (–0.61) (–0.46) (–0.34) Liquidity 0.057 0.055* 0.000 0.000 (1.59) (1.64) (0.01) (0.01) Institutional ownership –0.200* –0.220* –0.228*** –0.235*** (–1.63) (–1.82) (–2.69) (–3.09) Insider ownership –0.256 –0.272* –0.043 –0.038 (–1.58) (–1.77) (–0.59) (–0.85) Pill in force dummy 0.051 0.035 0.023 0.016 (0.88) (0.59) (0.34) (0.25) b$$_{\mathrm{1}}+$$b$$_{\mathrm{2}}$$–0.173* –0.056 (–1.84) (–0.63) b$$_{\mathrm{1}}+$$b$$_{\mathrm{2}}+$$b$$_{\mathrm{3}}$$0.031 0.100*** (0.45) (1.80) N 726 726 726 726 Includes year & industry dummies? Yes Yes Yes Yes Adjusted R$$^{\mathrm{2}}$$0.106 0.109 0.111 0.111 Buy and hold returns$$_{\mathrm{{[-1\ {\rm act}, +1\ {\rm exit}]}}}$$Holding-period CAR in excess of value-weighted CRSP index$$_{\mathrm{{[-1\ {\rm act}, +1\ {\rm exit}]}}}$$(1) (2) (3) (4) Resistance categorical variables Any resist (b$$_{\mathrm{1}})$$0.005 0.049* 0.042 0.029 (0.10) (1.65) (0.67) (0.36) Hostile resist (b$$_{\mathrm{2}})$$–0.222*** –0.086 (–2.86) (–0.99) Opposed hostile resist (b$$_{\mathrm{3}})$$0.205** 0.156*** (2.08) (2.50) Campaign characteristics Hedge fund offer dummy 0.277** 0.285** 0.175** 0.181** (1.98) (2.04) (1.95) (2.13) Other takeover dummy 0.089 0.087 0.053 0.049 (1.18) (1.16) (1.00) (0.89) Percentage owned –0.011** –0.010* –0.006 –0.006 (–2.19) (–1.89) (–1.24) (–1.10) Hedge fund in SEC group dummy 0.166 0.174 0.093* 0.103** (1.23) (1.27) (1.82) (2.11) Other 13D filer dummy –0.091 –0.088 –0.090*** –0.084*** (–1.50) (–1.42) (–3.16) (–2.74) Other explicit support dummy 0.008 0.012 0.004 0.004 (0.14) (0.21) (0.08) (0.09) Hedge fund characteristics Order quartile –0.009 –0.013 0.015 0.015 (–0.33) (–0.46) (0.43) (0.43) HF prior counterresistance dummy –0.010 –0.012 –0.034 –0.039 (–0.16) (–0.18) (–0.38) (–0.42) Firm characteristics Cash flow/assets –0.180 –0.165 –0.073 –0.065 (–0.64) (–0.62) (–0.28) (–0.26) log of market cap. ( \$$MM) –0.009 –0.006 –0.049*** –0.049** (–0.87) (–0.56) (–2.48) (–2.27) Cash –0.202 –0.168 –0.304* –0.290* (–0.94) (–0.79) (–1.85) (–1.80) Tobin’s q 0.021 0.022 0.027 0.026 (1.15) (1.15) (1.28) (1.25) RND 0.522 0.485 0.916*** 0.916*** (1.06) (1.03) (2.71) (2.84) Dividend yield –0.465 –0.538 –0.930 –1.029 (–0.27) (–0.33) (–0.94) (–1.13) Leverage –0.134* –0.137* –0.063 –0.064 (–1.87) (–1.94) (–0.68) (–0.69) HHI –0.181 –0.153 –0.084 –0.064 (–0.75) (–0.61) (–0.46) (–0.34) Liquidity 0.057 0.055* 0.000 0.000 (1.59) (1.64) (0.01) (0.01) Institutional ownership –0.200* –0.220* –0.228*** –0.235*** (–1.63) (–1.82) (–2.69) (–3.09) Insider ownership –0.256 –0.272* –0.043 –0.038 (–1.58) (–1.77) (–0.59) (–0.85) Pill in force dummy 0.051 0.035 0.023 0.016 (0.88) (0.59) (0.34) (0.25) b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}$$ –0.173* –0.056 (–1.84) (–0.63) b$$_{\mathrm{1}}+$$ b$$_{\mathrm{2}}+$$ b$$_{\mathrm{3}}$$ 0.031 0.100*** (0.45) (1.80) N 726 726 726 726 Includes year & industry dummies? Yes Yes Yes Yes Adjusted R$$^{\mathrm{2}}$$ 0.106 0.109 0.111 0.111 This table reports OLS regressions in which the dependent variable is either buy and hold returns (BHARs) (Columns 1 and 2) or holding-period CARs, both measured in excess of the value-weighted CRSP index$$_{\mathrm{[-1\ {\rm act},+1\ {\rm exit}]}}$$ (Columns 3 and 4). Means of exit include dropping below 5%; filing a 13G form; or being active in the firm when it delists, liquidates or declares bankruptcy, or merges. Regressions also include campaigns in which the hedge fund is still active in the target firm as of February 2017. For these latter campaigns, we calculate returns through February 28, 2017. All returns use daily data and include the period starting 1 day before the activism announcement and ending 1 day after the hedge fund exit. Any resist is set to 1 for the 356 campaigns with moderate or hostile target resistance and 0 otherwise. Hostile resist is set to 1 one for the subset of 207 resistance campaigns with hostile target resistance and 0 otherwise. Opposed hostile resist is set to 1 one for the subset of 138 campaigns with hostile target resistance and hedge fund counterresistance and 0 otherwise. The appendix describes all other control variables. Regressions include year and 1-digit SIC industry dummies. Standard errors are clustered by time and hedge fund, and t-values are reported below the coefficients in parentheses. HHI, Herfindahl-Hirschman index. ***, **, and * denotes statistical significance for the 1%, 5%, and 10% levels, respectively. Because Barber and Lyon (1997) argue convincingly that for long-term returns, BHARs are preferable to CARs; because our results for BHARs are more conservative; and because our results for BHARs and CARs are directionally similar, we focus our discussion on the BHARs in Table 8, Columns 1 and 2. Column 1 indicates that the BHARs for campaigns with any type of target resistance do not differ from the BHARs for campaigns with no resistance. Column 2 includes all three resistance categorical variables. The coefficients on these variables indicate that campaigns with moderate resistance outperform those with no resistance by about 5%, campaigns with unopposed hostile resistance perform worse than campaigns with moderate resistance by about 22%, and campaigns with opposed hostile resistance outperform campaigns with unopposed hostile resistance by about 21%. All three of these results are economically and statistically significant, and are consistent with the univariate results. The direction of the initial market reaction to hostile target firm resistance (negative) and the direction of the initial market reaction to formal hedge fund counterresistance (positive) correctly predicts the direction of long-term performance for each of these subsamples. However, the average net initial market return for the unopposed hostile resist and opposed hostile resist subsamples is less successful in predicting holding-period returns. For the unopposed hostile resist subsample, the negative 22% incremental effect of hostility from the target is sharply lower than the average positive net announcement return from activism and hostile target resistance of 2.8%. Similarly, for the opposed hostile resist subsample, the large 21% incremental effect of formal counterresistance by the hedge fund is sharply higher than the average positive net announcement return from activism, hostile target resistance, and formal hedge fund counterresistance of 1%. One possible explanation for this disparity is that there may be time variation in these results. However, when we examine the year-by-year effects, we do not detect any discernable patterns. A second possible explanation is that the average positive BHAR from activism is concentrated in cases in which target firms merge like in Boyson, Gantchev, and Shivdasani (2017) and that short-term CARs will be more successful in predicting activist holding-period returns for this subset of campaigns. Therefore, in unreported tests, we split the sample into two groups, those in which target firms merge within 18 months of the activism announcement and those in which they do not. Like in Boyson, Gantchev, and Shivdasani (2017), BHARs are significantly higher when firms merge; they are higher for the subsample with opposed hostile resistance (at 44.7%) compared to the subsample with unopposed hostile resistance (at 31.1%). Notably, net announcement period CARs are also higher, at 4.7% for opposed hostile resistance and 6% for unopposed hostile resistance. Examining firms in which targets do not merge within 18 months, the unopposed hostile resistance subsample has an average net announcement CAR of 2.2% and an average BHAR of (an insignificant) $$-$$11%, while for opposed hostile resistance the average net announcement CAR is 0.1% and the average BHAR is an insignificant 8%. Overall, it appears that for firms that merge in 18 months, the short-term net announcement returns are fairly predictive of long-term outcomes (i.e., the market appears to accurately predict the likelihood that the firm will merge). For firms that do not merge in 18 months, short-term net announcement returns seem to underreact to the target firm’s hostile resistance and do not predict BHARs for the subsample with unopposed hostile resistance, while they do predict BHARs for the sample with opposed hostile resistance. Returning to Table 8, Column 2, the sum of the resist and hostile resistance coefficients indicates that campaigns with unopposed hostile resistance significantly underperform campaigns with no resistance by about 17%. Further, the sum of all three coefficients implies that campaigns with formal hedge fund counterresistance perform similarly to campaigns with no resistance, consistent with an explanation that formal hedge fund counterresistance might mitigate the negative impact of hostile target firm resistance. To summarize, the modest positive impact of moderate resistance on holding-period returns documented in Table 8, Column 1, is consistent with the results for changes in cash flows. This finding supports the idea that moderate resistance may lead to more active negotiations with hedge funds, leading these firms to make proactive policy changes. At the very least, it appears that moderate target resistance does not erode hedge fund performance. Formal counterresistance is associated with the largest positive impact on holding-period returns. Campaigns with unopposed hostile resistance have holding periods that are worse than all the other subsamples and that are not significantly different from zero.28 Other than the subset of returns with unopposed hostile resistance, our results are consistent with prior literature that documents positive long-term returns to activism. Our key contributions to the long-term returns literature are to show that the market appears to initially under-react to both hostile target resistance and formal hedge fund counterresistance, and that unopposed hostile target firm resistance appears to negate the previously documented positive effect of hedge fund activism. 6. Conclusion Targets of hedge funds frequently resist activist overtures by worsening their corporate governance in hostile ways. Target firm managers are more likely to engage in hostile resistance when they have agency problems, when hedge funds express interest in buying the firm, and when institutional ownership is concentrated. These results imply that entrenched managers facing direct threats to their careers are more likely to engage in hostile resistance that inhibits shareholder voting power and reduces the ability for shareholders to enact changes by coordinating. Targets sometimes engage in less hostile (moderate) resistance by verbally denouncing hedge funds or making less aggressive governance changes. The market responds negatively to hostile resistance but is indifferent to moderate resistance. Unless hedge funds counterresist with proxy fights, unsolicited tender offers, or lawsuits, target firms have worse operating and long-term stock performance, a lower probability of mergers, and a higher probability of early hedge fund exit. The market positively responds to the announcement of hedge fund counterresistance, and counterresistance appears to partially offset the negative impact of hostile resistance. Campaigns with formal hedge fund counterresistance have target operating performance, merger probabilities, and short and long-term stock performance that do not significantly differ from campaigns with moderate target firm resistance or campaigns with no target firm resistance. Activist holding-period returns indicate that the market appears to underreact to both hostile target firm resistance and to formal hedge fund counterresistance, except in cases where the target firm eventually merges. Our finding of worse target firm outcomes in the face of unopposed hostile resistance does not appear to be driven by selection bias. We demonstrate numerous ways that hostile resistance directly constrains activists, from limits on ownership to restrictions on voting power. Further, the most significant differences in target firm performance occur when comparing unopposed hostile resistance campaigns to opposed hostile resistance campaigns. Our results have important implications for understanding the interactions between target firms and activists. The potential for resistance by entrenched target management should be considered by activists when targeting firms. Additionally, when activism is announced, existing target firm shareholders should consider the potentially negative implications of unopposed hostile resistance on future target firm performance. Appendix Table A1 Variable Descriptions Variable Description Data source Campaign characteristics Percentage owned Percentage of stock owned by the hedge fund during the campaign SharkRepellant, 13D filings Activist in SEC group dummy Dummy variable set to 1 if two or more hedge funds that jointly file a 13D form engage in activism SharkRepellant,13D filings Other 13D filer dummy Dummy variable set to 1 if the target firm has at least one other 13D filer active in the firm SharkRepellant, Proxy (SEC form DEF-14A), and 13D filings Other 13D percentage Total percentage owned by other 13D filers SharkRepellant, Proxy (SEC form DEF-14A), and 13D filings Other 13D filer has similar goal dummy Dummy set to 1 if the stated goal of the other 13D filers is similar to that of the current hedge fund activist, but these funds do not explicitly state support for each other SharkRepellant, Proxy (SEC form DEF-14A), 13D filings, and news searches (Factiva, Google News) Other explicit support dummy Dummy set to 1 if the other 13D filers explicitly state support for the current activist SharkRepellant, Proxy (SEC form DEF-14A), 13D filings, and news searches (Factiva, Google News) Hedge fund offer dummy Dummy set to 1 if the hedge fund makes an offer or expresses interest in making an offer for the firm SharkRepellant, 13D filings Other takeover dummy Dummy set to 1 if hedge fund states a goal that the firm be taken over, but the hedge fund does not make an offer SharkRepellant, 13D filings Hedge fund characteristics Order quartile Quartile of count of hedge fund manager’s activism events since 1994 up to but not including the event of interest (from Boyson, Ma, and Mooradian 2017); a higher value indicates more experience SharkRepellant and 13D filings HF prior counterresistance dummy Dummy variable set to one if the hedge fund has previously engaged in formal counterresistance against a target of activism SharkRepellant and 13D filings Target firm characteristics Cash flow/assets Cash flow scaled by average total assets Compustat ROA Return on assets Compustat Market cap Number of shares outstanding times stock price, in millions Compustat/CRSP Cash Cash scaled by average total assets Compustat Tobin’s q (book value of debt $$+$$ market value of common equity)/(book value of debt $$+$$ book value of common equity) Compustat RND Research and development expense scaled by average total assets Compustat Dividend yield Sum of preferred and common dividends scaled by the market value of common stock plus the book value of preferred stock Compustat/CRSP Leverage Book value of debt scaled by the book value of equity plus the book value of debt Compustat HHI Hirschman-Herfindahl index using 3-digit SIC codes Compustat Liquidity $$-$$1 (1$$+$$ln(Amihud x 10$$^{\mathrm{6}})$$; higher realizations imply higher liquidity.See Amihud (2002) CRSP Institutional ownership Estimated institutional ownership as a percentage of the total shares outstanding 13F filings and CRSP Shapley value Probability that a shareholder is the pivotal voter in a majority vote, summed across noninsider shareholders holding 3% or more 13F filings CEO tenure (years) Tenure of current CEO, in years Proxy (SEC form DEF-14A) or 10-K filings CEO is board chair dummy Dummy variable set to 1 if the CEO is the board chair Proxy (SEC form DEF-14A) Insider ownership Total director and officer percentage ownership Proxy (SEC form DEF-14A) or 10-K filings Pill in force dummy Dummy variable set to 1 if the target firm has a poison pill in force prior to the campaign Shark-Repellant and SEC form DEF-14A or 10-K filings Outcome variables Activism CAR$$_{[-1,+1]}$$ Cumulative abnormal returns (CARs), for the 3-day period beginning 1 day before activism is announced and ending 1 day after activism is announced. Calculated using a three-factor model including the market factor measured using the value-weighted CRSP index and the SMB and HML factors of Fama and French (1992, 1996). Following Greenwood and Schor (2009), loadings on these factors are calculated for the period 110 to 10 days before activism is announced CRSP, WRDS (for factors) Resistance CAR$$_{[-1,+1]}$$ Calculated using the same methodology as Activism CAR$$_{[-1,+1]}$$, where the event date is the date that target firm resistance is announced. For target firms that do not resist activists, we calculate a placebo CAR for the 3 days around the 14th day after activism is announced, since 14th days is the median time elapsed between the announcement of activism and the announcement of the resistance. The resistance CAR analysis excludes CARs for activism events in which the resistance announcement is confounded by another event during the 3-day period CRSP, WRDS (for factors) Buy and hold abnormal return (BHAR)$$_{[-1\ \textit{act},\ +1\ \textit{exit}]}$$ Buy and hold abnormal returns (BHARs) beginning 1 day before activism is announced and ending 1 day after the hedge fund exits the firm by reducing its stake to below 5%, switching to a 13G filing, or holding the stock when the firm merges or a firm merger is announced. Each day, the return on the stock in excess of the value-weighted CRSP return is calculated and the daily returns are then compounded over the entire holding period SEC form 13D, DEF-A or 10-K filings, news searches, CRSP for merger dates Holding-period CAR measured in excess of the value-weighted CRSP index$$_{[-1\ \textit{act},+1\ \textit{exit}]\ }$$ Cumulative abnormal returns (CARs) beginning 1 day before activism is announced and ending 1 day after the hedge fund exits the firm by reducing its stake to below 5%, switching to a 13G filing, or holding the stock when the firm merges or a firm merger is announced. Each day, the return on the stock in excess of the value-weighted CRSP return is calculated and the daily returns are then cumulated (added) over the entire holding period SEC form 13D, DEF-A or 10-K filings, news searches, CRSP for merger dates, stock prices, and index returns Acquired in 18 months dummy Dummy variable set to 1 if the target firm is acquired within 18 months of the announcement of activism CRSP, news searches (Factiva, Google News) Change in CF, 1 yr Percentage change in cash flows/assets for the year after activism relative to the year prior to activism Compustat Change in CF, 2 yr Percentage change in cash flows/assets 2 years after activism relative to the year prior to activism Compustat Change in ROA, 1 yr Percentage change in return on assets for the year after activism relative to the year prior to activism Compustat Change in ROA, 2 yr Percentage change in return on assets 2 years after activism relative to the year prior to activism Compustat Variable Description Data source Campaign characteristics Percentage owned Percentage of stock owned by the hedge fund during the campaign SharkRepellant, 13D filings Activist in SEC group dummy Dummy variable set to 1 if two or more hedge funds that jointly file a 13D form engage in activism SharkRepellant,13D filings Other 13D filer dummy Dummy variable set to 1 if the target firm has at least one other 13D filer active in the firm SharkRepellant, Proxy (SEC form DEF-14A), and 13D filings Other 13D percentage Total percentage owned by other 13D filers SharkRepellant, Proxy (SEC form DEF-14A), and 13D filings Other 13D filer has similar goal dummy Dummy set to 1 if the stated goal of the other 13D filers is similar to that of the current hedge fund activist, but these funds do not explicitly state support for each other SharkRepellant, Proxy (SEC form DEF-14A), 13D filings, and news searches (Factiva, Google News) Other explicit support dummy Dummy set to 1 if the other 13D filers explicitly state support for the current activist SharkRepellant, Proxy (SEC form DEF-14A), 13D filings, and news searches (Factiva, Google News) Hedge fund offer dummy Dummy set to 1 if the hedge fund makes an offer or expresses interest in making an offer for the firm SharkRepellant, 13D filings Other takeover dummy Dummy set to 1 if hedge fund states a goal that the firm be taken over, but the hedge fund does not make an offer SharkRepellant, 13D filings Hedge fund characteristics Order quartile Quartile of count of hedge fund manager’s activism events since 1994 up to but not including the event of interest (from Boyson, Ma, and Mooradian 2017); a higher value indicates more experience SharkRepellant and 13D filings HF prior counterresistance dummy Dummy variable set to one if the hedge fund has previously engaged in formal counterresistance against a target of activism SharkRepellant and 13D filings Target firm characteristics Cash flow/assets Cash flow scaled by average total assets Compustat ROA Return on assets Compustat Market cap Number of shares outstanding times stock price, in millions Compustat/CRSP Cash Cash scaled by average total assets Compustat Tobin’s q (book value of debt $$+$$ market value of common equity)/(book value of debt $$+$$ book value of common equity) Compustat RND Research and development expense scaled by average total assets Compustat Dividend yield Sum of preferred and common dividends scaled by the market value of common stock plus the book value of preferred stock Compustat/CRSP Leverage Book value of debt scaled by the book value of equity plus the book value of debt Compustat HHI Hirschman-Herfindahl index using 3-digit SIC codes Compustat Liquidity $$-$$1 (1$$+$$ln(Amihud x 10$$^{\mathrm{6}})$$; higher realizations imply higher liquidity.See Amihud (2002) CRSP Institutional ownership Estimated institutional ownership as a percentage of the total shares outstanding 13F filings and CRSP Shapley value Probability that a shareholder is the pivotal voter in a majority vote, summed across noninsider shareholders holding 3% or more 13F filings CEO tenure (years) Tenure of current CEO, in years Proxy (SEC form DEF-14A) or 10-K filings CEO is board chair dummy Dummy variable set to 1 if the CEO is the board chair Proxy (SEC form DEF-14A) Insider ownership Total director and officer percentage ownership Proxy (SEC form DEF-14A) or 10-K filings Pill in force dummy Dummy variable set to 1 if the target firm has a poison pill in force prior to the campaign Shark-Repellant and SEC form DEF-14A or 10-K filings Outcome variables Activism CAR$$_{[-1,+1]}$$ Cumulative abnormal returns (CARs), for the 3-day period beginning 1 day before activism is announced and ending 1 day after activism is announced. Calculated using a three-factor model including the market factor measured using the value-weighted CRSP index and the SMB and HML factors of Fama and French (1992, 1996). Following Greenwood and Schor (2009), loadings on these factors are calculated for the period 110 to 10 days before activism is announced CRSP, WRDS (for factors) Resistance CAR$$_{[-1,+1]}$$ Calculated using the same methodology as Activism CAR$$_{[-1,+1]}$$, where the event date is the date that target firm resistance is announced. For target firms that do not resist activists, we calculate a placebo CAR for the 3 days around the 14th day after activism is announced, since 14th days is the median time elapsed between the announcement of activism and the announcement of the resistance. The resistance CAR analysis excludes CARs for activism events in which the resistance announcement is confounded by another event during the 3-day period CRSP, WRDS (for factors) Buy and hold abnormal return (BHAR)$$_{[-1\ \textit{act},\ +1\ \textit{exit}]}$$ Buy and hold abnormal returns (BHARs) beginning 1 day before activism is announced and ending 1 day after the hedge fund exits the firm by reducing its stake to below 5%, switching to a 13G filing, or holding the stock when the firm merges or a firm merger is announced. Each day, the return on the stock in excess of the value-weighted CRSP return is calculated and the daily returns are then compounded over the entire holding period SEC form 13D, DEF-A or 10-K filings, news searches, CRSP for merger dates Holding-period CAR measured in excess of the value-weighted CRSP index$$_{[-1\ \textit{act},+1\ \textit{exit}]\ }$$ Cumulative abnormal returns (CARs) beginning 1 day before activism is announced and ending 1 day after the hedge fund exits the firm by reducing its stake to below 5%, switching to a 13G filing, or holding the stock when the firm merges or a firm merger is announced. Each day, the return on the stock in excess of the value-weighted CRSP return is calculated and the daily returns are then cumulated (added) over the entire holding period SEC form 13D, DEF-A or 10-K filings, news searches, CRSP for merger dates, stock prices, and index returns Acquired in 18 months dummy Dummy variable set to 1 if the target firm is acquired within 18 months of the announcement of activism CRSP, news searches (Factiva, Google News) Change in CF, 1 yr Percentage change in cash flows/assets for the year after activism relative to the year prior to activism Compustat Change in CF, 2 yr Percentage change in cash flows/assets 2 years after activism relative to the year prior to activism Compustat Change in ROA, 1 yr Percentage change in return on assets for the year after activism relative to the year prior to activism Compustat Change in ROA, 2 yr Percentage change in return on assets 2 years after activism relative to the year prior to activism Compustat Table A1 Variable Descriptions Variable Description Data source Campaign characteristics Percentage owned Percentage of stock owned by the hedge fund during the campaign SharkRepellant, 13D filings Activist in SEC group dummy Dummy variable set to 1 if two or more hedge funds that jointly file a 13D form engage in activism SharkRepellant,13D filings Other 13D filer dummy Dummy variable set to 1 if the target firm has at least one other 13D filer active in the firm SharkRepellant, Proxy (SEC form DEF-14A), and 13D filings Other 13D percentage Total percentage owned by other 13D filers SharkRepellant, Proxy (SEC form DEF-14A), and 13D filings Other 13D filer has similar goal dummy Dummy set to 1 if the stated goal of the other 13D filers is similar to that of the current hedge fund activist, but these funds do not explicitly state support for each other SharkRepellant, Proxy (SEC form DEF-14A), 13D filings, and news searches (Factiva, Google News) Other explicit support dummy Dummy set to 1 if the other 13D filers explicitly state support for the current activist SharkRepellant, Proxy (SEC form DEF-14A), 13D filings, and news searches (Factiva, Google News) Hedge fund offer dummy Dummy set to 1 if the hedge fund makes an offer or expresses interest in making an offer for the firm SharkRepellant, 13D filings Other takeover dummy Dummy set to 1 if hedge fund states a goal that the firm be taken over, but the hedge fund does not make an offer SharkRepellant, 13D filings Hedge fund characteristics Order quartile Quartile of count of hedge fund manager’s activism events since 1994 up to but not including the event of interest (from Boyson, Ma, and Mooradian 2017); a higher value indicates more experience SharkRepellant and 13D filings HF prior counterresistance dummy Dummy variable set to one if the hedge fund has previously engaged in formal counterresistance against a target of activism SharkRepellant and 13D filings Target firm characteristics Cash flow/assets Cash flow scaled by average total assets Compustat ROA Return on assets Compustat Market cap Number of shares outstanding times stock price, in millions Compustat/CRSP Cash Cash scaled by average total assets Compustat Tobin’s q (book value of debt $$+$$ market value of common equity)/(book value of debt $$+$$ book value of common equity) Compustat RND Research and development expense scaled by average total assets Compustat Dividend yield Sum of preferred and common dividends scaled by the market value of common stock plus the book value of preferred stock Compustat/CRSP Leverage Book value of debt scaled by the book value of equity plus the book value of debt Compustat HHI Hirschman-Herfindahl index using 3-digit SIC codes Compustat Liquidity $$-$$1 (1$$+$$ln(Amihud x 10$$^{\mathrm{6}})$$; higher realizations imply higher liquidity.See Amihud (2002) CRSP Institutional ownership Estimated institutional ownership as a percentage of the total shares outstanding 13F filings and CRSP Shapley value Probability that a shareholder is the pivotal voter in a majority vote, summed across noninsider shareholders holding 3% or more 13F filings CEO tenure (years) Tenure of current CEO, in years Proxy (SEC form DEF-14A) or 10-K filings CEO is board chair dummy Dummy variable set to 1 if the CEO is the board chair Proxy (SEC form DEF-14A) Insider ownership Total director and officer percentage ownership Proxy (SEC form DEF-14A) or 10-K filings Pill in force dummy Dummy variable set to 1 if the target firm has a poison pill in force prior to the campaign Shark-Repellant and SEC form DEF-14A or 10-K filings Outcome variables Activism CAR$$_{[-1,+1]}$$ Cumulative abnormal returns (CARs), for the 3-day period beginning 1 day before activism is announced and ending 1 day after activism is announced. Calculated using a three-factor model including the market factor measured using the value-weighted CRSP index and the SMB and HML factors of Fama and French (1992, 1996). Following Greenwood and Schor (2009), loadings on these factors are calculated for the period 110 to 10 days before activism is announced CRSP, WRDS (for factors) Resistance CAR$$_{[-1,+1]}$$ Calculated using the same methodology as Activism CAR$$_{[-1,+1]}$$, where the event date is the date that target firm resistance is announced. For target firms that do not resist activists, we calculate a placebo CAR for the 3 days around the 14th day after activism is announced, since 14th days is the median time elapsed between the announcement of activism and the announcement of the resistance. The resistance CAR analysis excludes CARs for activism events in which the resistance announcement is confounded by another event during the 3-day period CRSP, WRDS (for factors) Buy and hold abnormal return (BHAR)$$_{[-1\ \textit{act},\ +1\ \textit{exit}]}$$ Buy and hold abnormal returns (BHARs) beginning 1 day before activism is announced and ending 1 day after the hedge fund exits the firm by reducing its stake to below 5%, switching to a 13G filing, or holding the stock when the firm merges or a firm merger is announced. Each day, the return on the stock in excess of the value-weighted CRSP return is calculated and the daily returns are then compounded over the entire holding period SEC form 13D, DEF-A or 10-K filings, news searches, CRSP for merger dates Holding-period CAR measured in excess of the value-weighted CRSP index$$_{[-1\ \textit{act},+1\ \textit{exit}]\ }$$ Cumulative abnormal returns (CARs) beginning 1 day before activism is announced and ending 1 day after the hedge fund exits the firm by reducing its stake to below 5%, switching to a 13G filing, or holding the stock when the firm merges or a firm merger is announced. Each day, the return on the stock in excess of the value-weighted CRSP return is calculated and the daily returns are then cumulated (added) over the entire holding period SEC form 13D, DEF-A or 10-K filings, news searches, CRSP for merger dates, stock prices, and index returns Acquired in 18 months dummy Dummy variable set to 1 if the target firm is acquired within 18 months of the announcement of activism CRSP, news searches (Factiva, Google News) Change in CF, 1 yr Percentage change in cash flows/assets for the year after activism relative to the year prior to activism Compustat Change in CF, 2 yr Percentage change in cash flows/assets 2 years after activism relative to the year prior to activism Compustat Change in ROA, 1 yr Percentage change in return on assets for the year after activism relative to the year prior to activism Compustat Change in ROA, 2 yr Percentage change in return on assets 2 years after activism relative to the year prior to activism Compustat Variable Description Data source Campaign characteristics Percentage owned Percentage of stock owned by the hedge fund during the campaign SharkRepellant, 13D filings Activist in SEC group dummy Dummy variable set to 1 if two or more hedge funds that jointly file a 13D form engage in activism SharkRepellant,13D filings Other 13D filer dummy Dummy variable set to 1 if the target firm has at least one other 13D filer active in the firm SharkRepellant, Proxy (SEC form DEF-14A), and 13D filings Other 13D percentage Total percentage owned by other 13D filers SharkRepellant, Proxy (SEC form DEF-14A), and 13D filings Other 13D filer has similar goal dummy Dummy set to 1 if the stated goal of the other 13D filers is similar to that of the current hedge fund activist, but these funds do not explicitly state support for each other SharkRepellant, Proxy (SEC form DEF-14A), 13D filings, and news searches (Factiva, Google News) Other explicit support dummy Dummy set to 1 if the other 13D filers explicitly state support for the current activist SharkRepellant, Proxy (SEC form DEF-14A), 13D filings, and news searches (Factiva, Google News) Hedge fund offer dummy Dummy set to 1 if the hedge fund makes an offer or expresses interest in making an offer for the firm SharkRepellant, 13D filings Other takeover dummy Dummy set to 1 if hedge fund states a goal that the firm be taken over, but the hedge fund does not make an offer SharkRepellant, 13D filings Hedge fund characteristics Order quartile Quartile of count of hedge fund manager’s activism events since 1994 up to but not including the event of interest (from Boyson, Ma, and Mooradian 2017); a higher value indicates more experience SharkRepellant and 13D filings HF prior counterresistance dummy Dummy variable set to one if the hedge fund has previously engaged in formal counterresistance against a target of activism SharkRepellant and 13D filings Target firm characteristics Cash flow/assets Cash flow scaled by average total assets Compustat ROA Return on assets Compustat Market cap Number of shares outstanding times stock price, in millions Compustat/CRSP Cash Cash scaled by average total assets Compustat Tobin’s q (book value of debt $$+$$ market value of common equity)/(book value of debt $$+$$ book value of common equity) Compustat RND Research and development expense scaled by average total assets Compustat Dividend yield Sum of preferred and common dividends scaled by the market value of common stock plus the book value of preferred stock Compustat/CRSP Leverage Book value of debt scaled by the book value of equity plus the book value of debt Compustat HHI Hirschman-Herfindahl index using 3-digit SIC codes Compustat Liquidity $$-$$1 (1$$+$$ln(Amihud x 10$$^{\mathrm{6}})$$; higher realizations imply higher liquidity.See Amihud (2002) CRSP Institutional ownership Estimated institutional ownership as a percentage of the total shares outstanding 13F filings and CRSP Shapley value Probability that a shareholder is the pivotal voter in a majority vote, summed across noninsider shareholders holding 3% or more 13F filings CEO tenure (years) Tenure of current CEO, in years Proxy (SEC form DEF-14A) or 10-K filings CEO is board chair dummy Dummy variable set to 1 if the CEO is the board chair Proxy (SEC form DEF-14A) Insider ownership Total director and officer percentage ownership Proxy (SEC form DEF-14A) or 10-K filings Pill in force dummy Dummy variable set to 1 if the target firm has a poison pill in force prior to the campaign Shark-Repellant and SEC form DEF-14A or 10-K filings Outcome variables Activism CAR$$_{[-1,+1]}$$ Cumulative abnormal returns (CARs), for the 3-day period beginning 1 day before activism is announced and ending 1 day after activism is announced. Calculated using a three-factor model including the market factor measured using the value-weighted CRSP index and the SMB and HML factors of Fama and French (1992, 1996). Following Greenwood and Schor (2009), loadings on these factors are calculated for the period 110 to 10 days before activism is announced CRSP, WRDS (for factors) Resistance CAR$$_{[-1,+1]}$$ Calculated using the same methodology as Activism CAR$$_{[-1,+1]}$$, where the event date is the date that target firm resistance is announced. For target firms that do not resist activists, we calculate a placebo CAR for the 3 days around the 14th day after activism is announced, since 14th days is the median time elapsed between the announcement of activism and the announcement of the resistance. The resistance CAR analysis excludes CARs for activism events in which the resistance announcement is confounded by another event during the 3-day period CRSP, WRDS (for factors) Buy and hold abnormal return (BHAR)$$_{[-1\ \textit{act},\ +1\ \textit{exit}]}$$ Buy and hold abnormal returns (BHARs) beginning 1 day before activism is announced and ending 1 day after the hedge fund exits the firm by reducing its stake to below 5%, switching to a 13G filing, or holding the stock when the firm merges or a firm merger is announced. Each day, the return on the stock in excess of the value-weighted CRSP return is calculated and the daily returns are then compounded over the entire holding period SEC form 13D, DEF-A or 10-K filings, news searches, CRSP for merger dates Holding-period CAR measured in excess of the value-weighted CRSP index$$_{[-1\ \textit{act},+1\ \textit{exit}]\ }$$ Cumulative abnormal returns (CARs) beginning 1 day before activism is announced and ending 1 day after the hedge fund exits the firm by reducing its stake to below 5%, switching to a 13G filing, or holding the stock when the firm merges or a firm merger is announced. Each day, the return on the stock in excess of the value-weighted CRSP return is calculated and the daily returns are then cumulated (added) over the entire holding period SEC form 13D, DEF-A or 10-K filings, news searches, CRSP for merger dates, stock prices, and index returns Acquired in 18 months dummy Dummy variable set to 1 if the target firm is acquired within 18 months of the announcement of activism CRSP, news searches (Factiva, Google News) Change in CF, 1 yr Percentage change in cash flows/assets for the year after activism relative to the year prior to activism Compustat Change in CF, 2 yr Percentage change in cash flows/assets 2 years after activism relative to the year prior to activism Compustat Change in ROA, 1 yr Percentage change in return on assets for the year after activism relative to the year prior to activism Compustat Change in ROA, 2 yr Percentage change in return on assets 2 years after activism relative to the year prior to activism Compustat We thank two anonymous referees; the editor; Raj Aggarwal, Aaron Brauner, Rudi Fahlenbrach, Erik Lie, Elgar Pichler, Jeff Pontiff, Luke Taylor, and Ralph Walkling; and seminar participants at the 2016 European Finance Association Annual Meeting, the Oxford Conference for Corporate Reputation, Brandeis University, and Northeastern University. Footnotes 1Keusch (2017) shows that CEOs of activist targets suffer job loss and reputational damage. Brav, Jiang, and Kim (2015) find that while productivity improves at target firms, workers relinquish much of this value to equity investors. Klein and Zur (2009) show that hedge funds often gain board seats, and Greenwood and Schor (2009) find that activist targets are more likely to merge. 2 See Brav, Jiang, and Kim (2015) and Becht et al. (2017). 3 The E index is based on 6 of the 24 governance provisions that comprise the GIM index. Four of the provisions limit shareholder voting power: limits to shareholder bylaw amendments, supermajority voting provisions for mergers or charter amendments, and classified boards. Two additional provisions—poison pills and golden parachutes—are often enacted in anticipation of hostile offers. See Bebchuk, Cohen, and Ferrell (2009). 4 We also classify target firm lawsuits and bylaw changes limiting shareholder ability to call special meetings or act by written consent as hostile. Because hedge funds often rely on the support of other investors (see Appel, Gormley, and Keim 2016; Becht et al. 2017; Brav, Dasgupta, and Mathews 2017), these bylaw changes can impede activists. 5Brav, Dasgupta, and Mathews (2017), Appel, Gormley, and Keim (2016), and Becht et al. (2017) discuss the relation between hedge fund activists and other institutional investors, and Chakraborty and Gantchev (2013) find that more concentrated ownership improves coordination among institutional investors. 6 Having more cash makes firm management more vulnerable to agency problems (Stulz 1990); high ownership concentration makes firms more vulnerable to investor coordination (Chakraborty and Gantchev 2013); and CEO duality and long CEO tenure can indicate entrenchment (see, e.g., Jensen 1986; Goyal and Park 2002; Masulis, Wang, and Xie 2007). 7 While we model and examine formal hedge fund counterresistance in response to hostile target firm resistance, we do not explicitly examine a hedge fund’s response to moderate target resistance. In the majority of these campaigns, hedge funds respond to the moderate target resistance by negotiating with the target firm. These negotiations do not meet the threshold of hostile target firm resistance, and they do not vary significantly from one another along any observable dimensions. 8 Consistent with the results of Brav et al. (2008), our untabulated results show that in the full sample, hedge funds achieve their stated goals of activism about 50% of the time. 9 A large literature argues that the threat of passive institutional investors selling their shares (“exiting”) can have a disciplinary effect on management, especially when a stock is more liquid (Edmans 2009; Edmans and Manso 2011; Bharath, Jayaraman, and Nagar 2013; Edmans, Fang, and Zur 2013). In contrast to passive investors, activist investors also have the option to intervene (“voice”). Hence, the activist might choose to intervene in a less liquid stock to avoid a potential negative price impact on eventual sale (Bhide 1993; Coffee 1991; Bharath, Jayaraman, and Nagar 2013; Edmans, Fang, and Zur 2013). 10 The direction of the separate short-term announcement returns to resistance and counterresistance is accurately reflected in activist holding-period returns (i.e., holding-period returns are lower for unopposed hostile target resistance and higher for opposed hostile target resistance, consistent with the announcement returns). However, the net short-term CARs do not differ for the unopposed and opposed hostile resist subsamples, and therefore do not accurately predict the future differences in activist holding-period returns. Specifically, the sum of the short-term CARs for activism, resistance, and counterresistance is 2.8% for the unopposed hostile resist subsample and 1% for the opposed hostile resist subsample. However, the holding-period performance results indicate that the unopposed hostile resist sample has a long-term return that is not different from zero, whereas the opposed hostile resist sample has a holding-period return that is significantly better than zero and significantly higher than the return for the unopposed hostile resist sample. Hence, it appears that the market underreacts to both the announcement of hostile resistance and the announcement of hedge fund counterresistance. We discuss these results further in Section 6. 11 Although it is possible that these two subsets of campaigns might differ among unobservable dimensions, we control for numerous observable variables in performing these analyses. 12 In the hostile resistance subsample, over half of campaigns involving poison pills restrict an activist’s additional purchasing power to less than 2% of the target firm’s market capitalization. Other direct constraints include limits to special meetings enacted after the activist has expressed the intent to call a meeting and board classification in response to activists seeking board seats. 13 See Brav et al. (2008), Klein and Zur (2009), Greenwood and Schor (2009), Boyson, Gantchev, and Shivdasani (2017), Clifford (2008), Becht et al. (2009), and Brav, Jiang, and Kim (2015). 14 SR data include Form 13D filings and other purposeful campaigns. An investor with activist intentions who crosses the 5% ownership threshold must file a Form 13D with the SEC, stating the purpose of the campaign. About half of all 13D campaigns are purposeful (consistent with Brav et al. 2008). We review the entire campaign to determine whether activism is purposeful. SR includes other campaigns in which the activist owns less than 5% of the target firm and does not file a Form 13D, such as proxy fights, lawsuits, or unsolicited tender offers. Since these campaigns are purposeful, we include them in our sample. 15Ryngaert (1988) and Malatesta and Walkling (1988) report negative stock returns around poison pill announcements. Del Guercio and Hawkins (1999) document negative short-term stock returns around antitakeover proposal announcements (including but not limited to poison pills). 16Thomas and Cotter (2007) find that proposals to change or remove poison pills are frequently supported by majority votes. 17Bebchuk et al. (2013) find that 76% of pills in effect in 2012 had thresholds of 15% or less and 15% had thresholds of 10% or less, in contrast with previous studies documenting 20% or, even, 30% thresholds for earlier time periods. 18 For a case study in which another activist states explicit support for the hedge fund activist, consider Metropolitan Capital Advisers and Cyberonics, Inc. On January 4, 2007, Metropolitan, a 7.3% stakeholder, initiated a proxy fight to elect three members to the eight-member board of Cyberonics. Metropolitan had been campaigning for board representation since September 2006. On January 22, 2007, Carl Icahn announced a 9.8% stake and stated his support for Metropolitan’s proxy fight. On January 29, 2007, Metropolitan and Cyberonics settled the proxy fight, appointing Metropolitan’s three nominees to the board. In the year (2 years) following activism, Cyberonics had a 39% (68%) improvement in cash flows. 19 The BHARs are geometric returns using daily data and calculated like in Barber and Lyon (1997), whereas the CARs are additive. As with all other summary statistics in Table 1, the average returns are equally weighted. If the hedge fund has not exited the campaign by February 2017, the return is calculated through February 2017. Our average holding period is 2.7 years. Our 11.7% average BHAR for the full sample is consistent with prior studies of hedge fund activism. Bebchuk, Brav, and Jiang (2015) report a 7.2% average BHAR for the period beginning the second month after activism and ending 3 years later. They report 4% returns in the first month after activism, for a total 3-year BHAR of 11.2%, which is nearly the same as our average BHAR of 11.7% for a similar holding period, despite using a different time frame (1993–2007). Our annualized 4.3% BHAR (11.7% holding period/2.7 years) is very close to the annualized holding-period return of 4.8% reported in Boyson, Gantchev, and Shivdasani (2017), who use the same time frame (2001–2012) that we do. Finally, our return is somewhat lower than that in Brav et al. (2008), who report an annualized BHAR of 20% for the time from 2001 to 2006. 20 The 0.9% counterresistance CAR is in line with prior literature. Most hedge fund counterresistance is in the form of proxy contests. Fos (2016, figure 3) indicates a return of about 1% around the announcement of proxy contests. 21 We exclude campaigns for which we infer hostile target resistance, since the interpretation of the coefficient on prior formal hedge fund counterresistance is complicated by the inclusion of these campaigns. Specifically, for this subset, the coefficient on prior formal counterresistance measures the relation between prior formal counterresistance and future formal counterresistance. Doing so muddies its interpretation in the full sample. As a robustness test, we reperform the regressions with the full sample of target resistance, excluding the prior formal counterresistance independent variable. Results on the other coefficients do not change, indicating that our key results are not driven by observed hostile resistance. 22 We thank an anonymous referee for this suggestion. 23Hedge fund offer dummy includes several cases in which hedge funds make unsolicited tender offers. Since an unsolicited tender offer is one type of formal hedge fund counterresistance, this variable will be mechanically correlated with the dependent variable. 24 In untabulated results we also examine the incidence of hedge fund board representation and find that activists achieve board representation in 33% of campaigns. This proportion rises to 51% for campaigns with formal hedge fund counterresistance (which includes many proxy fights). Board representation does not significantly vary across the other samples. Since we consider board representation to be an intermediate step and not the ultimate goal of activism, we do not explore it further. 25 In this and all remaining analyses, we do not include the Shapley value, CEO tenure, and CEO is board chair variables because data are missing for about 100 campaigns. In unreported analyses, we include these variables in the analyses and find that they are rarely significant and never change the signs or significance of coefficients on the resistance categorical variables. We also interact the Shapley value with both unopposed hostile resistance and opposed hostile resistance (since we have shown that targets (hedge funds) are more (less) likely to engage in hostile resistance (oppose hostile resistance) when ownership is more concentrated). However, this interaction term is never significant. 26 Since encouraging a merger is a common goal of hedge fund activism, it is not surprising that many hedge funds exit after the completion of a merger. Examining the 8% of cases (62 campaigns) in which target firms remain active through the end of the sample period, we note that these cases are dominated by GAMCO and also include activists such as Steel Partners, Wynnefield Partners, and Carl Icahn. These cases often include board representation for the hedge fund or its representatives (46% of the time) and a high hedge fund ownership stake (about 14%). Activists in these campaigns appear to take a very long-term view. 27 In the multivariate (and univariate) analyses, we cannot calculate long-term returns for 95 campaigns, primarily because we cannot precisely determine the hedge fund exit date. 28 In unreported tests, we show that the holding-period returns to hedge funds in this subsample that exit within 1 year of activism without achieving their stated activism goals do not differ from the holding-period returns to campaigns in this subsample in which hedge funds do not exit early. References Admati, A. R. , and Pfleiderer P. . 2009 . The “Wall Street walk” and shareholder activism: Exit as a form of voice . Review of Financial Studies 22 : 2645 – 85 . Google Scholar CrossRef Search ADS Amihud, Y. 2002 . Illiquidity and stock returns: Cross-section and time-series effects . Journal of Financial Markets 5 : 31 – 56 . Google Scholar CrossRef Search ADS Appel, I. R. , Gormley T. A. , and Keim D. B. . 2016 . Standing on the shoulders of giants: The effect of passive investors on activism. Working Paper , Boston College and Wharton. Barber, B. M. , and Lyon J. D. . 1997 . “Detecting long-run abnormal stock returns: The empirical power and specification of test statistics. Journal of Financial Economics 43 : 341 – 72 . Google Scholar CrossRef Search ADS Bebchuk, L. A. , Brav, A. Jackson, R. J. Jr., and Jiang. W. 2013 . Pre-disclosure accumulations by activist investors: Evidence and policy . Journal of Corporation Law 39 : 1 – 34 . Bebchuk, L. A. , Brav A. , and Jiang W. . 2015 . The long-term effects of hedge fund activism . Columbia Law Review 115 : 1085 – 155 . Bebchuk, L. A. , Cohen A. , and Ferrell A. . 2009 . What matters in corporate governance? Review of Financial Studies 22 : 783 – 827 . Google Scholar CrossRef Search ADS Becht, M. , Franks J. , Mayer C. , and Rossi S. . 2009 . Returns to shareholder activism: Evidence from a clinical study of the Hermes UK Focus Fund . Review of Financial Studies 22 : 3093 – 129 . Google Scholar CrossRef Search ADS Becht, M. , Franks J. , Grant J. , and Wagner H. F. . 2017 . The returns to hedge fund activism: An international study . Review of Financial Studies 30 : 2933 – 71 . Google Scholar CrossRef Search ADS Bharath, S. T. , Jayaraman S. , and Nagar V. . 2013 . Exit as governance: An empirical analysis , Journal of Finance 68 : 2525 – 47 . Google Scholar CrossRef Search ADS Bhide, A. 1993 . The hidden costs of stock market liquidity. Journal of Financial Economics 34 : 31 – 51 . Google Scholar CrossRef Search ADS Boyson, N. , Gantchev N. , and Shivdasani A. . 2017 . Merger activism. Journal of Financial Economics 126 : 54 – 73 . Google Scholar CrossRef Search ADS Boyson, N. , Ma L. , and Mooradian R. . 2017 . Serial activists. Working Paper , Northeastern University. Google Scholar CrossRef Search ADS Brav, A. , Dasgupta A. , and Mathews R. . 2017 . Wolf pack activism. Working Paper , Duke University . Google Scholar CrossRef Search ADS Brav, A. , Jiang W. , and Kim H. . 2015 . The real effects of hedge fund activism: Productivity, asset allocation, and labor outcomes . Review of Financial Studies 28 : 2723 – 69 . Google Scholar CrossRef Search ADS Brav, A. , Jiang W. , Partnoy F. , and Thomas R. . 2008 . Hedge fund activism, corporate governance, and firm performance . Journal of Finance 63 : 1729 – 75 . Google Scholar CrossRef Search ADS Chakraborty, I. , and Gantchev N. . 2013 . Does shareholder coordination matter? Evidence from private placements. Journal of Financial Economics 108 : 213 – 30 . Google Scholar CrossRef Search ADS Chemmanur, T. , Paeglis I. , and Simonyan K. . 2011 . Management quality and antitakeover provisions . Journal of Law and Economics 54 : 651 – 92 . Google Scholar CrossRef Search ADS Coffee, J. C. 1991 . Liquidity versus control: The institutional investor as corporate monitor . Columbia Law Review 91 : 1277 – 368 . Google Scholar CrossRef Search ADS Clifford, C. 2008 . Value creation or destruction? Hedge funds as shareholder activists . Journal of Corporate Finance 14 : 323 – 36 . Google Scholar CrossRef Search ADS Cremers, M. , and Ferrell A. . 2014 . Thirty years of shareholder rights and firm value . Journal of Finance 69 : 1167 – 96 . Google Scholar CrossRef Search ADS DeAngelo, H. , and Rice E. . 1983 . Antitakeover charter amendments and stockholder wealth . Journal of Financial Economics 11 : 329 – 59 . Google Scholar CrossRef Search ADS Del Guercio, D. , and Hawkins J. . 1999 . The motivation and impact of pension fund activism . Journal of Financial Economics 52 : 293 – 340 . Google Scholar CrossRef Search ADS Edmans, A. 2009 . Blockholder trading, market efficiency, and managerial myopia . Journal of Finance 64 : 2481 – 513 . Google Scholar CrossRef Search ADS Edmans, A. , Fang V. W. , and Zur E. . 2013 . The effect of liquidity on governance . Review of Financial Studies 26 : 1443 – 82 . Google Scholar CrossRef Search ADS Edmans, A. , and Manso G. . 2011 . Governance through trading and intervention: A theory of multiple blockholders . Review of Financial Studies 24 : 2395 – 28 . Google Scholar CrossRef Search ADS Fama, E. F. , and French K. R. . 1992 . The cross-section of expected stock returns . Journal of Finance 47 : 427 – 65 . Google Scholar CrossRef Search ADS Fama, E. F. , and French K. R. . 1996 . Multifactor explanations of asset pricing anomalies . Journal of Finance 51 : 55 – 84 . Google Scholar CrossRef Search ADS Fos, V. 2016 . The disciplinary effects of Proxy contests . Management Science 63 : 655 – 71 . Google Scholar CrossRef Search ADS Gantchev, N. 2013 . The costs of shareholder activism: Evidence from a sequential decision model. Journal of Financial Economics 107 : 610 – 31 . Google Scholar CrossRef Search ADS George, B. , and Lorsch J. W. . 2014 . How to outsmart activist investors. Harvard Business Review , May . https://hbr.org/2014/05/how-to-outsmart-activist-investors. Gillan, S. L. , and Starks L. T. . 2000 . Corporate governance proposals and shareholder activism: the role of institutional investors. Journal of Financial Economics 57 : 275 – 305 . Google Scholar CrossRef Search ADS Gompers, P. , Ishii J. , and Metrick A. . 2003 . Corporate governance and equity prices . Quarterly Journal of Economics 118 : 107 – 56 . Google Scholar CrossRef Search ADS Goyal, V. K. , and Park C. W. . 2002 . Board leadership structure and CEO turnover . Journal of Corporate Finance 8 : 49 – 66 . Google Scholar CrossRef Search ADS Greenwood, R. , and Schor M. . 2009 . Investor activism and takeovers . Journal of Financial Economics 92 : 362 – 75 . Google Scholar CrossRef Search ADS Heron, R. , and Lie E. . 2015 . The effect of poison pill adoptions and court rulings on firm entrenchment . Journal of Corporate Finance 35 : 286 – 96 . Google Scholar CrossRef Search ADS Jensen, M. C. 1986 . Agency costs of free cash flow, corporate finance, and takeovers . American Economic Review 76 : 323 – 29 . Jiang, W. , Li T. , and Mei D. . Forthcoming . Influencing control: Jawboning in risk arbitrage. Journal of Finance . Keusch, T. 2017 . Hedge fund activists and CEO contracting. Working Paper, INSEAD . Kini, O. , Kracaw W. , and Mian S. . 2004 . The nature of discipline by corporate takeovers . Journal of Finance 59 : 1511 – 55 . Google Scholar CrossRef Search ADS Klein, A. , and Zur E. . 2009 . Entrepreneurial shareholder activism. Hedge funds and other private investors . Journal of Finance 64 : 187 – 229 . Google Scholar CrossRef Search ADS Kothari, S. P. , and Warner J. B. . 2007 . Econometrics of event studies. In Handbook of corporate finance: Empirical corporate finance , Volume 1 , pp. 1 – 36 . Ed. Eckbo. B. E. Amsterdam : Elsevier . Google Scholar CrossRef Search ADS Lipton, M. 2012 . Dealing with activist hedge funds. Harvard Law School Forum on Corporate Goverance and Financial Regulation, August 9 . http://clsbluesky.law.columbia.edu/2013/06/21/dealing-with-activist-hedge-funds/. Malatesta, P. H. , and Walkling R. A. . 1988 . Poison pill securities: Stockholder wealth, profitability and ownership structure . Journal of Financial Economics 20 : 347 – 76 . Google Scholar CrossRef Search ADS Masulis, R. W. , Wang C. and Xie. F. 2007 . Corporate governance and acquirer returns . Journal of Finance 62 : 1851 – 89 . Google Scholar CrossRef Search ADS Maug, E. 1998 . Large shareholders as monitors: Is there a trade-off between liquidity and control? Journal of Finance 53 : 65 – 98 . Google Scholar CrossRef Search ADS Milnor, J. W. , and Shapley L. S. . 1978 . Values of large games II: Oceanic games . Mathematics of Operations Research 3 : 290 – 307 . Google Scholar CrossRef Search ADS Ryngaert, M. 1988 . The effect of poison pill securities on shareholder wealth . Journal of Financial Economics 20 : 377 – 417 . Google Scholar CrossRef Search ADS Safieddine, A. , and Titman S. . 1999 . Leverage and corporate performance: evidence from unsuccessful takeovers . Journal of Finance 54 : 547 – 80 . Google Scholar CrossRef Search ADS Stulz, R. M. 1990 . Managerial discretion and optimal financing policies . Journal of Financial Economics 26 : 3 – 27 . Google Scholar CrossRef Search ADS Thomas, R. , and Cotter J. . 2007 . Shareholder proposals in the new millennium: Shareholder support, board response, and market reaction . Journal of Corporate Finance 13 : 368 – 91 . Google Scholar CrossRef Search ADS © The Author(s) 2018. Published by Oxford University Press on behalf of The Society for Financial Studies. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Review of Financial Studies Oxford University Press

# Hostile Resistance to Hedge Fund Activism

, Volume Advance Article – May 17, 2018
47 pages