* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Successful Activist Board Engagement
Survey
Document related concepts
Syndicated loan wikipedia , lookup
Beta (finance) wikipedia , lookup
Business valuation wikipedia , lookup
Private equity wikipedia , lookup
Modified Dietz method wikipedia , lookup
Investment fund wikipedia , lookup
Stock selection criterion wikipedia , lookup
Global saving glut wikipedia , lookup
Early history of private equity wikipedia , lookup
Private equity secondary market wikipedia , lookup
Private equity in the 1980s wikipedia , lookup
Financial economics wikipedia , lookup
Fund governance wikipedia , lookup
Shareholder value wikipedia , lookup
Corporate finance wikipedia , lookup
Transcript
Myopic Investor Myth Debunked: Shareholder Advocacy The Long-term Efficacy of Hedge Fund Activism in the Boardroom Shane Goodwin Spears School of Business Oklahoma State University Ramesh Rao Spears School of Business Oklahoma State University Abstract Over the past two decades, hedge fund activism has emerged as new form of corporate governance mechanism that brings about operational, financial and governance reforms to a corporation. Many prominent business executives and legal scholars are convinced that the American economy will suffer unless hedge fund activism with its perceived short-termism agenda is significantly restricted. Shareholder activists and their proponents claim they function as a disciplinary mechanism to monitor management and are instrumental in mitigating the agency conflict between managers and shareholders. We find statistically meaningful empirical evidence to reject the anecdotal conventional wisdom that hedge fund activism is detrimental to the long term interests of companies and their long term shareholders. Moreover, our findings suggest that hedge funds generate substantial long term value for target firms and its long term shareholders when they function as a shareholder advocate to monitor management through active board engagement. JEL Classifications: D21, D22, D81, G12, G23, G32, G34, G35, G38, K22 I. INTRODUCTION For society as a whole, further empowering [hedge funds] with short-term holding periods subjects Americans to lower long-term growth and job creation…due to excessive risk taking…when corporations maximize short-term profits. Leo Strine, Chief Justice of the Delaware Supreme Court Columbia Law Review, March 2014 Leo Strine, the Chief Justice of the Delaware Supreme Court, the country’s most important arbiter of corporate law recently authored a 54-page article in the Columbia Law Review with respect to his views about hedge funds and his belief that short-term investors are detrimental to the long-term interests of American corporations, its long-term shareholders and to society as a whole. Chief Justice Strine, many prominent business executives and legal scholars are convinced that the American economy will suffer unless hedge fund activism with its perceived short-termism agenda is significantly restricted. Shareholder activists and their proponents claim they function as a disciplinary mechanism to monitor management and are instrumental in mitigating the agency conflict between managers and shareholders. In May 2014, the Delaware Court of Chancery denied Third Point, a well-known activist hedge fund, the ability to increase its ownership above 10% in Sotheby’s. Third Point, the company’s largest shareholder, sent a letter to Sotheby’s CEO in October 2013 expressing concerns regarding the company’s strategic direction, shareholder misalignment and recommended replacing the CEO. In response, Sotheby’s adopted a unique two-tier shareholder rights plan1 (a poison pill) that purposely discriminated and severely punished any shareholder that might have an agenda that conflicts with incumbent management2. Although the court 1 A shareholder rights plan, also known as a “poison pill”, is one of the most effective defense tactics available to publicly traded corporations. A poison pill is designed to make a potential transaction extremely unattractive to a hostile party from an economic perspective, compelling a suitor to negotiate with the target’s Board of Directors. A poison pill is effective because it dilutes the economic interest of the hostile suitor in the target, making the transaction both economically unattractive and impractical if pursued on a hostile basis. 2 The two-tier structure provided for a 10 percent trigger threshold for shareholders filing a Schedule 13D (for “active” investors) and a 20 percent trigger for shareholders filing a Schedule 13G (available to “passive” investors). The rights plan also contained a “qualifying offer” exception, which provided that the plan would not apply to an offer for all of Sotheby’s shares and expires in one year unless it is approved by a shareholder vote. 1 viewed the two-tier trigger structure as “discriminatory” — differentiating between activist and passive investors — the court found that Sotheby’s had a reasonable basis to believe that Third Point posed a threat of exercising disproportionate control and influence over major decisions. The court reached its decision notwithstanding the support from other large shareholders for Third Point and despite the fact that Sotheby’s management and board owned less than 1% of the company. Ironically and inexplicably, the court’s decision de facto authorized the incumbent CEO and the board disproportionate control and influence over major decisions, notwithstanding the objections of its shareholders. Vice Chancellor Donald Parsons, Jr. stated in his opinion “…it’s important not to overstate the way in which shareholders that file Schedule 13Ds differ from those who file Schedule 13Gs.” The rationale for this decision coupled with the recent remarks by Chief Justice Strine raise important policy questions about the value of hedge fund activism and its disciplinary role as an active monitor of a firm’s management. In this paper, we suggest and empirically test the following alternative hypothesis: hedge fund activism through board representation is not detrimental to the long term operating performance of companies and does not have an adverse effect on the target firm’s long term shareholders. Agency conflict in publicly traded corporations with dispersed ownership is at the heart of corporate governance literature, which focuses on mechanisms to discipline incumbent management. One possible solution to mitigate agency cost is for shareholders to actively monitor the firm’s management. However, while monitoring may reduce agency and improve firm value, this effort is not without cost and the benefits from monitoring are enjoyed by all shareholders (Grossman and Hart, 1980). Shareholders that serve as active monitors of firm management to provide a disciplinary mechanism is not a new concept. Earlier studies show that when institutional investors, particularly mutual funds and pension funds, follow an activist agenda, they do not achieve 2 significant benefits for shareholders (Black (1998), Karpoff (2001), Romano (2001), and Gillan and Starks (2007)). However, hedge funds have increasingly engaged in shareholder activism and monitoring that differs fundamentally from previous activist efforts by other institutional investors. Unlike mutual funds and pension funds, hedge funds are able to influence corporate boards and managements due to key differences arising from their organizational form and the incentive structures. Hedge funds employ highly incentivized managers who manage large unregulated pools of capital. Because they are not subject to regulation that governs mutual funds and pension funds, they can hold highly concentrated positions in a small number of companies, and use leverage and derivatives to extend their reach. In addition, hedge fund managers don’t experience conflicts of interest since they are not beholden to the management of the firms whose shares they hold. In sum, hedge funds are better positioned to act as informed monitors than other institutional investors. The growing literature with respect to shareholder activism identifies a significant positive stock price reaction for targeted companies with the announcement of an activist intervention ((Brav, Jiang and Kim, 2009), Clifford (2008) and Boyson and Mooradian (2011)). Many critics of hedge fund activism concede that there are short-term positive value increases to the target firm and its shareholders as a result of self-interested “myopic investors”. While this “myopic investor” claim has been regularly invoked and has had considerable influence, its supporters, including Chief Justice Strine, have thus far failed to support their position with empirical evidence. However, the continued debate is about the long term efficacy on target firms and the returns to all shareholders as result of hedge fund activism. Recent research by Bebchuk, Brav and Jiang (2013) find statistically meaningful evidence that the operating performance of target firms improves following activist interventions but no evidence to support the claim that short-term improvement was at the expense of long-term performance. 3 The dataset used by Bebchuk, Brav and Jiang (2013) is consistent with the vast majority of research with respect to shareholder activism. The sample of activist interventions is primarily constructed from Schedule 13D filings, the mandatory federal securities law filings under Section 13(d) of the 1934 Exchange Act. The law states that investors must file with the SEC within 10 days of acquiring more than 5% of any class of securities of a publicly traded company if they have an interest in influencing the management of the company. The presumption is a shareholder that files a 13D is unequivocally motivated to change the strategic direction of the company. However, we claim that is not always the case. In fact, some activist campaigns are centered on corporate governance reforms (i.e., board declassification, removal of shareholder rights plan, etc.) and not meaningful long-term strategic changes to the target firm. We contend that any shareholder with sincere conviction to challenge the current strategic direction of a firm would, ultimately, seek board representation if their demands were not supported by the firm’s incumbent management. We view board representation as a signal of an activist’s long term commitment to the firm. Additionally, board representation is a costly endeavor that is not borne by all shareholders which further validates the activist’s credibility as a long term shareholder of the firm. The vast majority of shareholder activism literature is predicated on Schedule 13D filings. However, we assert that the optimal dataset to empirically test the long-term effects of shareholder activism should be based on board representation of target firms by a shareholder activist. Therefore, we started with a much more expansive sample of activist interventions. Figure I illustrates our comprehensive dataset of shareholder activist events, which includes 5,063 interventions from 1984-2013. Of those, 3,899 (77%) filed a 13D. However, approximately 32% of all activist interventions were focused on board engagement, either through a proxy contest (1,216) or a non-proxy contest dissident campaign that resulted in board representation via private negotiations (418) with the target management team and its board of 4 directors. To be sure, over two-thirds (2/3) of activist interventions did not seek board representation to actively monitor management. Moreover, GAMCO Asset Management, a hedge fund founded by Mario Gabeli, has filed 478 13Ds since 1996. However, it has launched only 18 proxy fights (3.8%) and won board representation only ten times (2.1%) to date. In contrast, Carl Icahn has launched proxy fights and won board representation at eBay, Genzyme, Time Warner and Yahoo! without filing a 13D. Accordingly, we claim that there are numerous 13D filings of activist interventions that otherwise include good performing companies with strong management that a dissident was not compelled to seek board representation to actively monitor management and function as a disciplinary mechanism. Additionally, there are over 90 activist interventions that led to board representation without filing a 13D. Therefore, we assert that the optimal dataset to test empirically the long-term effects of hedge fund activism should be based on board representation of target firms by a shareholder activist and not merely the fact that a shareholder crossed 5% ownership and might (not will) seek to influence strategic change at the target firm. To be sure, an activist that is willing to incur significant financial cost that is not borne by all shareholders, which Gantchev (2012) estimates is approximately $10 million per proxy contest, has genuine conviction that the target firm requires strategic change that management is unwilling to execute without shareholder interference. We empirically test our manually constructed dataset of 448 activist interventions (the “Treatment Group”) that resulted in at least one board seat granted to an activist hedge fund from 1996-2013 (see Table 1). A total of 843 board members (see Table 1) were elected at 398 unique target companies. This includes 225 unique activist hedge funds. Of the 448 activist interventions in the Treatment Group, 243 (54%) target firms are still publicly-listed, 186 (42%) were sold/merged and 19 (4%) target firms filed for bankruptcy. 5 By compiling our own database, we avoid some problems associated with survivorship bias, reporting selection bias, and backfill, which are prevalent among other hedge fund databases. To control for self-selection bias and endogeneity, we constructed a control group (the “Control Group”) of all proxy fights campaigns that did not result in board representation during the same period. Our dataset includes 595 target firms that experienced a proxy contest for board representation. After we excluded certain events to reflect consistent sample parameters with our Treatment Group, our Control Group includes 73 firms that were involved in a proxy contest that the target firm incumbent management defeated the dissident shareholder during the voting process. Therefore, we examined not only firms that granted at least one board seat to a dissident shareholder and its ex post effects (the “Treatment Group”), but also companies that were challenged by dissatisfied shareholders and did not suffer the ex post disciplinary effects by the dissident (the “Control Group”). During our investigation of abnormal returns during the review period, we employed three standard methods used by financial economists for detecting stock return performance. In particular, the study examines: first, whether the returns to targeted companies were systematically lower than what would be expected given standard asset pricing models; second, whether the returns to targeted companies were lower than those of the Control Group that experienced a similar event; and, third, whether a portfolio based on taking positions in activism targets and holding them for five years post the board seat grant date underperforms relative to its risk characteristics. Additionally, we modeled an 18-year (1996-2013) buy-andhold abnormal returns (BHAR) portfolio of all shareholder activist interventions that resulted in board representation and controlled for market risk, firm size and value tilt relative to the Control Group. Using the aforementioned financial and econometric models, we find no evidence that target firms experience a “reversal of fortune” during the five-year period following the 6 intervention. The long-term underperformance asserted by supporters of the myopic activism claim, and the resulting losses to long-term shareholders due to activist interventions, are not found in the data. Moreover, we find target firms that granted at least one board seat to an activist hedge fund created positive abnormal returns (alpha) for all shareholders during short term event windows and for a five year period ex post the activist joining the target firm board. Additionally, those target firms increased certain operating performance measures that are commonly used by financial economists, such as return on assets (ROA) and market value relative to book value (Tobin’s Q) during the post event period. Further, the assertion that myopically-focused activist investors only create value in the short-term at the expense of longterm shareholders is not supported by the data. In fact, target firms that granted at least one board seat to an activist hedge fund outperformed the Control Group with respect to firm operating measures and positive abnormal returns for all shareholders during the five years ex post the activist joining the board. We find that target firms in our sample dataset underperformed their industry peers on certain operating metrics prior to the activist intervention, which validates the value-oriented characteristics of activist targets. However, those operating metrics progressively improved during the review period subsequent to the board seat grant date. This is further evidence that board representation by activist hedge funds lead to improved operating performance in the long term. In contrast, target firms that won the proxy fight against the activist experienced degradation in certain operating performance metrics compared to industry peers during the review period. Contrary to certain extant literature and to the prevailing narrative that activist hedge funds frequently promote a “sell the company” agenda, we find that activists who seek board representation do not promote such an objective. However, within two years of the dissident 7 shareholder joining the board, the CEO of the target firm had been replaced approximately 30% of the time. Therefore, it is reasonable to assume from the data that hedge funds that seek board representation are focused more on promoting change at the target firm via operating improvements and changes to existing management rather than supporting a sale of the company strategy. Our investigation and findings support the alternative hypothesis that hedge fund activism is not detrimental to and does not have an adverse-effect on the long term interests of target firms and their long term shareholders (see Graph 1 and Graph 2). Our research fills the important void with respect to the long term efficacy of shareholder activists serving as a disciplinary mechanism on the firm by actively seeking board representation to monitor management. Additionally, we contribute to the literature regarding shareholder activists as self-interested myopic investors at the expense of the long-term interest of the company and its long term shareholders. Moreover, our findings have important policy implications related to the ongoing debate on corporate governance and the rights and roles of shareholders. Although some prominent legal commentators and presiding justices, such as Chief Justice Strine, have called for restrictions on hedge fund activism because of its perceived short-term orientation, our findings suggest that hedge fund activism generates substantial long term value for target firms and its long term shareholders when they function as a shareholder advocate to monitor management through active board engagement. 8 II. LITERATURE REVIEW In this section, we review the extant literature with respect to shareholder activism. First, we evaluate why shareholder activists are the optimal group to be effective monitors of firm management to mitigate agency cost. Second, we will discuss why active board engagement is the preferred path to create value rather than through passive shareholder proposals. Finally, we will discuss how shareholders can intervene to effect change at a corporation via a proxy contest. The separation of ownership and control in public firms gives rise to the possibility of agency conflict between the firm’s managers and shareholders (Berle and Means (1932) and Jensen and Meckling (1976)). To the extent that this agency cost is significant, it can have a detrimental effect on shareholder value. The agency problem in publicly traded corporations with dispersed ownership is at the heart of corporate governance literature, which focuses on mechanisms to discipline incumbent management. One possible solution to mitigate agency cost is for shareholders to actively monitor the firm’s management. However, while monitoring may reduce agency and improve firm value, this effort is not without cost and the benefits from monitoring are enjoyed by all shareholders (Grossman and Hart, 1980). Shareholders that serve as active monitors of firm management to provide a disciplinary mechanism is not a new concept. Gillan and Starks (2007) define shareholder activists as “investors who, dissatisfied with some aspect of a company’s management or operations, try to bring about change within the company without a change in control.” Tirole (2006) provides the following definition: “Active monitoring consists in interfering with management in order to increase the value of the investors’ claims.” However, hedge funds have increasingly engaged in shareholder activism and monitoring that differs fundamentally from previous activist efforts by other institutional investors. Earlier 9 studies show that when institutional investors, particularly mutual funds and pension funds, follow an activist agenda, they do not achieve significant benefits for shareholders (Black (1998), Karpoff (2001), Romano (2001), and Gillan and Starks (2007)). Unlike mutual funds and pension funds, hedge funds are able to influence corporate boards and managements due to key differences arising from their organizational form and the incentive structures. Hedge funds employ highly incentivized managers who manage large unregulated pools of capital. Because they are not subject to regulation that governs mutual funds and pension funds, they can hold highly concentrated positions in a small number of companies, and use leverage and derivatives to extend their reach. In addition, hedge fund managers don’t experience conflicts of interest since they are not beholden to the management of the firms whose shares they hold. In sum, hedge funds are better positioned to act as informed monitors than other institutional investors. Theory predicts that large shareholders should be effective monitors of the managers of publicly listed firms, reducing the free-rider problem ((Shleifer and Vishny (1986) and Grossman and Hart (1980)). Yet the evidence that large shareholders increase shareholder value is mixed. In two recent surveys, Karpoff (2001) and Romano (2001) conclude that activism conducted by large institutional shareholders (i.e., pension funds and mutual funds) has had little impact on firm performance. Additionally, Karpoff, Malatesta, and Walkling (1996), Wahal (1996), and Gillan and Starks (2000) report no persuasive evidence that shareholder proposals increase firm values, improve operating performance or even influence firm policies. Therefore, hedge funds are the best positioned to function as a shareholder advocates to monitor management through active board engagement. Brav, Jiang, Partnoy, and Thomas (2008) find that the announcement of hedge fund 10 activism generates abnormal returns of more than 7% in a short window around the announcement. In addition, the authors document modest changes in operating performance around the activism. Klein and Zur (2009) and Clifford (2007) also document significant positive abnormal returns around the announcement of activism. Recent research by Bebchuk, Brav and Jiang (2013) find statistically meaningful evidence that the operating performance of target firms improves following activist interventions but no evidence to support the claim that short-term improvement was at the expense of long-term performance. When shareholders are dissatisfied with the performance of a corporation and its’ board of directors, they can intervene via a proxy contest. The proxy contest process is a meticulously regulated election mechanism which can be invoked when “one group, referred to as ‘dissidents’ or ‘insurgents’ attempt to obtain seats on the firm’s board of directors currently in the hands of another group, referred to as ‘incumbents’ or ‘management’” (Dodd and Warner, 1983). The objective is to displace incumbents with the dissidents’ preferred candidates in order to bring about an overall improvement in enterprise financial performance and shareholder value. Although dissident shareholders do not always obtain a majority of board seats, in many cases they manage to capture some seats. Notwithstanding proxy contest outcome, there is evidence that share price performance is positively and significantly associated with the proxy contest process (Dodd and Warner, 1983). Within three years of a proxy contest event, many publicly held firms experience major changes including resignations of top management within one year of the contest followed by sale or liquidation of the firm. Proxy contest shareholder gains derive largely from the dissident linked sale of the corporation (DeAngelo and DeAngelo, 1989). These findings are consistent with our investigation. Within two years of a dissident shareholder joining the board of a target 11 firm, the CEO resigned approximately 30% of the time and over the course of the five years 21% of our target firms were sold/merged. Mulherin and Poulsen (1998) determined that “on average, proxy contests create value.” Research confirms that the bulk of shareholder wealth gains arise from firms that are acquired in the period surrounding the contest. In contrast, management turnover in firms that are not acquired results in a significant and positive effect on stock owners’ value proposition because organizations engaged in management change out are more inclined to re-structure following a proxy contest event. The rate of management turnover for proxy contest challenged firms is much higher compared to organizations not involved in proxy contest activity and is directly proportional to the share of seats at the board won by proxy contenders. When the majority of seats are won by proxy contesters, the highest management turnover is observed reflecting the importance of intangible issues such as job security (Yen and Chen, 2005). Bebchuk, Brav and Jiang (2013) found that contrary to the claim that hedge fund activists adversely impact the long-term interests of organizations and their shareholders, there is evidence that activist interventions lead to improved operating performance in the five years that follow the interventions. Venkiteshwaran, Iyer and Rao (2010) conducted a detailed study of hedge fund activist Carl Icahn’s 13D filings and subsequent firm performance and found significant share price increases for the target companies (of about 10%) around the time Icahn discloses his intentions publicly. Additionally, the author’s found a significant number (1/3) of Icahn’s targets ended up being acquired or taken private within 18 months of his initial investment. The shareholders of those companies earned abnormal returns of almost 25% from the time of Icahn’s initial investment through the sale of the company. This finding is consistent 12 with the DeAngelo and DeAngelo (1989) research that shareholder gains derive largely from the dissident linked sale of the corporation. Our research fills the important void with respect to the long term efficacy of shareholder activists serving as a disciplinary mechanism on the firm by actively seeking board representation to monitor management. Additionally, we contribute to the literature regarding shareholder activists as self-interested myopic investors at the expense of the long-term interest of the company and its long term shareholders. Moreover, our findings have important policy implications related to the ongoing debate on corporate governance and the rights and roles of shareholders. Our findings suggest that hedge fund activism generates substantial long term value for target firms and its long term shareholders when they function as a shareholder advocate to monitor management through active board engagement. 13 III. DATA AND METHODOLOGY There is no central database of activist hedge funds. Therefore, we constructed an independent dataset of all activist interventions from 1984-2013 from various sources, including Compustat, Capital IQ, FactSet, ISS Proxy Data, SharkRepellent and the SEC’s EDGAR database. Additionally, our dataset includes Schedule 13D filings, the mandatory federal securities law filings under Section 13(d) of the 1934 Exchange Act that investors must file with the SEC within 10 days of acquiring more than 5% of any class of securities of a publicly traded company if they have an interest in influencing the management of the company. Our manually constructed database of shareholder activist events includes 5,063 interventions from 1984-2013 (see Figure I). Similar to Gillan and Starks (2007), we define shareholder activist event as a purposeful intervention by “investors who, dissatisfied with some aspect of a company’s management or operations, try to bring about change within the company without a change in control.” Our data collection comprised a multi-step procedure. The vast majority of shareholder activism literature is predicated on Schedule 13D filings. However, we assert that the optimal dataset to empirically test the long-term effects of shareholder activism should be based on board representation of target firms by a shareholder activist. Therefore, we started with a much more expansive sample of activist interventions. Our comprehensive dataset of shareholder activist events includes 5,063 interventions from 1984-2013. Of those, 3,899 (77%) filed a 13D. However, approximately 32% of all activist interventions were focused on board engagement, either through a proxy contest (1,216) or non-proxy contest dissident campaigns that resulted in board representation via private negotiations (418) with the target management team and board of directors. To be sure, over two-thirds (2/3) of activist 14 interventions did not seek board representation to actively monitor management. Moreover, GAMCO Asset Management, a hedge fund founded by Mario Gabeli, has filed 478 13Ds since 1996. However, it has launched only 18 proxy fights (3.8%) and won board representation only ten times (2.1%) to date. In contrast, Carl Icahn has launched proxy fights and won board representation at eBay, Genzyme, Time Warner and Yahoo! without filing a 13D. Accordingly, we claim that there are numerous 13D filings of activist interventions that otherwise include good performing companies with strong management that an activist is not compelled to seek board representation to actively monitor management and function as a disciplinary mechanism. Additionally, there are over 90 activist interventions that led to board representation without filing a 13D. Therefore, we assert that the optimal dataset to empirically test the long-term effects of hedge fund activism should be based on board representation of target firms by a shareholder activist and not merely the fact that a shareholder crossed 5% ownership and might (not will) seek to influence strategic change at the target firm. To be sure, an activist that is willing to incur significant financial cost that is not borne by all shareholders, which Gantchev (2012) estimates is approximately $10 million per proxy contest, has genuine conviction that the target firm requires strategic change that management is unwilling to execute. In our second step, we narrowed our time-frame from 1996-2013 and identified 1,039 activist interventions that resulted in board representation either through a proxy fight or private negotiations. This sample set included 621 proxy fights and 418 activist interventions (nonproxy contests) that resulted in board representation either through a settlement or concessions between the target management and the dissident shareholder. Next, we excluded certain events where: (1) the primary purpose of the filer is to be involved in the bankruptcy 15 reorganization or the financing of a distressed firm; and (2) the target is a closed-end fund or other non-regular corporation. We excluded duplicate campaigns by multiple activists (i.e., the wolf-pack) so the dataset includes information about the target firm only once with respect to a specific campaign. If a target firm were to file for bankruptcy protection or liquidation, we included financial information from the target firm up to the Chapter 11 or Chapter 7 filing date. More specifically, the bankrupt firm would account for 100% loss as it relates to stock return and portfolio analyses. In our final step, we included only hedge fund activist campaigns at target firms with a market capitalization of $50 million or greater. Although, there are many campaigns that are targeted at micro-cap companies, we determined that the trading liquidity and financial data of these firms were not representative of the broader sample. Additionally, we trimmed certain variables and financial data at the 1.0% and 99.0% in each tail to adjust for outliers. Our final dataset consists of 448 activist interventions (the “Treatment Group”) that resulted in at least one board seat granted to an activist shareholder from 1996-2013 (see Table 1). A total of 843 board members were elected at 398 unique target companies. This includes 225 unique activist hedge funds. Of the 448 activist interventions in the Treatment Group, 243 (54%) target firms are still publicly-listed, 186 (42%) were sold/merged and 19 (4%) target firms filed for bankruptcy. By compiling our own database, we avoid some problems associated with survivorship bias, reporting selection bias, and backfill, which are prevalent among other hedge fund databases. Table 4 provides descriptive statistics with respect to the Treatment Group. 16 Table 2 Company Status Existing Sold/Merged Bankruptcy Total 243 186 19 448 54% 42% 4% Primary Campaign Type Board Representation Board Control Maximize Shareholder Value No Publicly Disclosed Activism Other Activist Campaigns Total 235 69 80 40 24 448 52% 15% 18% 9% 5% Proxy Fight Winner Settled/Concessions Made 174 Dissident 53 Split 14 Total 241 Activism Type Proxy Fight Other Stockholder Campaign No Publicly Disclosed Activism Exempt Solicitation 241 165 40 2 Total 448 72% 22% 6% 54% 37% 9% 0% As previously noted, extant literature determined that approximately 1/3 of certain activist interventions led to a sale or merger of the target firm within 18 months of the intervention. However, we find that hedge fund activists that seek board representation do not promote a “sell the company” agenda consistent with other activist objectives. Table 3 illustrates that less than 7% of the target firms announced a sale of the company ex post the dissident board seat grant date. Additionally, Table 3 highlights CEO changes at target firms post a dissident joining the board after an activist campaign. Within two years of the shareholder activist joining the board, the CEO had been replaced approximately 30% of the time. Therefore, it is reasonable to assume from the data that hedge funds that seek board representation are focused more on promoting change at the target firm via operating improvements and changes to existing management rather than supporting a sale of the company strategy. 17 Table 3 Announced Target Firm Sale or Divestiture of Assets Ex Post Dissident Seat Date 3-5 Years No. of Target Firms 19 % of Total Board Engagements 4.2% 2-3 Years 17 1-2 Years CEO Change Ex Post Dissident Seat Date 3-5 Years No. of Target Firms 36 % of Total Board Engagements 8.0% 3.8% 2-3 Years 27 6.0% 27 6.0% 1-2 Years 46 10.3% 6-12 months 14 3.1% 6-12 months 37 8.3% < Six (6) Months 17 3.8% < Six (6) Months 40 8.9% Total 94 Total 186 Timeframe Timeframe To control for self-selection bias and endogeneity, we constructed a control group (the “Control Group”) of all proxy fights campaigns that did not result in board representation during the same period. Our dataset includes 595 target firms that experienced a proxy contest for board representation. After we excluded certain events to reflect consistent sample parameters with our Treatment Group, our Control Group includes 73 firms that were involved in a proxy contest that the target firm incumbent management defeated the dissident shareholder during the voting process. Therefore, we examined not only firms that granted at least one board seat to a dissident shareholder and its ex post effects (the “Treatment Group”), but also companies that were challenged by dissatisfied shareholders and did not suffer the ex post disciplinary effects by the dissident (the “Control Group”). Table 4 provides an overview of certain characteristics the target firms and the respective shareholder activist tactics with respect to each intervention for the Treatment Group (N=448) and the Control Group (N=73). 18 Table 4 Descriptive Statistics of Dataset Treatment Group: Target Firms that Granted at least One Board Seat (N=448) Target Firm Characteristics Classified Board Cumulative Voting Poison Pill Adopted in Response to Campaign Poison Pill In Force Prior to Announcement Control Group: Target Firm Management Won Proxy Fight (No Shareholder Activist Board Representation, N=73) Yes 210 No 238 Yes (%) 47% No (%) 53% Yes 35 No 38 Yes (%) 48% No (%) 52% 34 33 414 415 8% 7% 92% 93% 3 3 70 70 4% 4% 96% 96% 174 274 39% 61% 30 43 41% 59% Shareholder Activist Tactics 13D Filer 400 48 89% 11% 47 26 64% 36% Dissident Group Includes SharkWatch50 Member 261 187 58% 42% 38 35 52% 48% Dissident Tactic: Nominate Slate of Directors 260 188 58% 42% 60 13 82% 18% Dissident Tactic: Publicly Disclosed Letter to Board/Management 248 200 55% 45% 50 23 68% 32% Proxy Fight 241 207 54% 46% 73 0 100% 0% Standstill Agreement Special Exhibit Included 135 313 30% 70% 1 72 1% 99% Publicly Disclosed Letter to Management 129 53 71% 29% 26 7 79% 21% Dissident Seek Reimbursement 120 328 27% 73% 39 34 53% 47% Dissident Tactic: Letter to Stockholders 115 333 26% 74% 61 12 84% 16% Proxy Fight Went Definitive 114 334 25% 75% 73 0 100% 0% Dissident's Fight Letter Special Exhibit Included 87 361 19% 81% 48 25 66% 34% Short Slate 82 366 18% 82% 36 37 49% 51% Dissident Tactic: Request Company Seek Buyer 75 373 17% 83% 13 60 18% 82% Shareholder Proposals (Excluding Dissident Director Nominees) 71 377 16% 84% 24 49 33% 67% Proxy Fight Went the Distance 67 381 15% 85% 73 0 100% 0% Letter to Shareholders 66 116 36% 64% 30 3 91% 9% Dissident Tactic: Threaten Proxy Fight 41 407 9% 91% 0 73 0% 100% Dissident Tactic: Lawsuit 39 409 9% 91% 15 58 21% 79% Dissident Tactic: Propose Binding Proposal 39 409 9% 91% 8 65 11% 89% Dissident Tactic: Propose Precatory Proposal 32 416 7% 93% 12 61 16% 84% Dissident Tactic: Remove Director(s) 27 421 6% 94% 5 68 7% 93% Dissident Tactic: Unsolicited Offer 19 429 4% 96% 2 71 3% 97% Threaten Proxy Fight 18 164 10% 90% 0 33 0% 100% Hostile or Unsolicited Offer 14 168 8% 92% 0 33 0% 100% Dissident Tactic: Remove Officer(s) 14 434 3% 97% 2 71 3% 97% Special Meeting 13 435 3% 97% 10 63 14% 86% Dissident Tactic: Take Action by Written Consent 12 436 3% 97% 0 73 0% 100% Dissident Tactic: Call Special Meeting 11 437 2% 98% 4 69 5% 95% Written Consent 11 437 2% 98% 0 73 0% 100% 20 IV. EMPIRICAL TESTS AND RESULTS In this section, we address and empirically test the alternative hypothesis that hedge fund activism is not detrimental to the long term operating performance of target firms or its long term shareholders. To test the ex post disciplinary effects by dissident shareholders we examined and measured firm performance and long term abnormal stock returns up to five years from the board seat grant date. Additionally, we conducted similar tests during the five year period prior to the intervention. We contrast those results with both Control Groups. a. Measuring Cross-Sectional Firm Operating Performance We measure firm performance by several empirical proxies: Tobin’s Q, returns on assets (ROA), return on equity (ROE), return on invested capital (ROIC) and stock returns, which are the most widely used and accepted firm performance proxies. Additionally, since the conventional wisdom and the common narrative is that shareholder activists negatively affect long term operating performance and decrease capital investment, we measured operating margin (EBIT/Sales) and capital investment policy (CAPEX/Sales) of the target firms for the Treatment Group and for both Control Groups. We report the industry adjusted differences for ROA and Tobin’s Q for each target sample firm in the Treatment Group and both Control Groups compared to similar matched firms (firms in the same four-digit SIC industry). Tobin’s Q is named after the Nobel Prize winner James Tobin and is calculated as the ratio of market value to asset replacement value (Yermack, 1996). Tobin’s Q is calculated as: Tobin’s Q= (Market value of assets) / (Replacement cost of assets) As an approximation, the market value of assets is computed as market value of equity plus book value of assets minus book value of equity, following Brown and Caylor (2006). The 21 asset replacement value is taken as the book value of assets. A Tobin’s Q ratio greater than one indicates the good quality of a firms investment decisions: it has invested in positive NPV investment projects rather than in negative NPV investment projects and the returns meet or exceed expectations. In contrast, Tobin’s Q lower than one suggests that the firm did not even earn its returns expected from investors from the investment projects to cover the cost of capital. Return on assets is calculated as earnings before interests, and taxes (EBIT) multiplied by the reciprocal of the effective rate divided by the average of total assets for the year. Return on assets (ROA) indicates how efficient management is at using its assets to generate earnings. Calculated by dividing a company’s annual earnings by its total assets, ROA is generally displayed as a percentage. Sometimes this is referred to as “return on investment”, an indicator of how profitable a company is: Return on assets (ROA) = (EBIT * (1-tax rate)) / ((Total Assets(t) + Total Assets(t-1)) / 2) Return on invested capital (ROIC) is calculated as earnings before interest, and taxes (EBIT) multiplied by the reciprocal of the effective rate divided by the average of total debt and total common equity for the target firms. ROIC calculation is used to assess a firm’s efficiency at allocating the capital under its control to profitable investments. The return on invested capital measure gives a sense of how well a company is using its money to generate returns. The calculation is as follows: ROIC = (EBIT * (1-tax rate)) / ((Total Debt + Preferred Equity + Total Common Equity)(t)) + ((Total Debt + Preferred Equity + Total Common Equity)(t-1)) / 2) Return on equity (ROE) is calculated as earnings after tax and interest (net income) divided by the average of total common equity and total preferred equity plus minority interest 22 for the target firms. ROE calculation is used to assess a firm’s profitability by revealing how much profit a company generates with the capital shareholders have invested. The calculation is as follows: ROE = Net Income / ((Total Common Equity + Preferred Equity + Total Minority)) Operating margin is calculated as earnings before interest and taxes divided by total revenue generated by the target firms. A ratio used to measure a company’s pricing strategy and operating efficiency. Operating margin is a measurement of what proportion of a company’s revenue remains after paying for variable costs of production such as wages, raw materials, etc. and provides an understanding of how much a company makes (before interest and taxes) on each dollar of sales. The calculation is as follows: Operating Margin = EBIT / Sales Capital expenditures relative to sales (CAPEX / Sales) is calculated as total capital expenditures divided by the firm’s revenues. This metric provides measure of how much the firm is investing for future growth opportunities as well as maintaining existing fixed assets. The calculation is as follows: CAPEX / Sales = Total Capital Expenditures / Total Revenues Our findings of target firm performance for companies that granted at least one board seat to a dissident shareholder are presented in Panel A in Table 5 and our findings of target firm performance for both Control Groups are presented in Panel B and Panel C in Table 5. Consistent with the extant literature (Brav et al, 2008), we find that target firms’ operating performance for the Treatment Group and both Control Groups deteriorates prior to an activist intervention (defined as the “Event”). For example, the median ROA for the Treatment 23 Group declines significantly from 3.46% five years prior to the Event date to 1.72% on the date the board seat is granted to a dissident shareholder. The foregoing results were statistically significant at the 1% level. The median ROA for the Treatment Group increased from 1.72% on the grant seat date to 3.17% five years post the Event date, yielding a 13% CAGR increase during the review period (see Graph 3). However, during the same period the Control Group experienced a significant decline in ROA post the Event date (the announcement date of the proxy contest results). Graph 3 Return on Assets (ROA) 4.50 4.00 3.50 Board Seat Grant Date All Target Firms that granted at least one board seat to a dissident: 13% CAGR increase in ROA over 5 year period 3.53 3.17 3.00 2.50 2.00 1.72 All Target Firms that WON the proxy fight against activist hedge fund: 18% CAGR 1.50 decrease in ROA over 5 year period 1.35 1.00 Event 1 +1 2 +2 3 +3 4 +4 5 6 +5 Years The mean Q Ratio for the Treatment Group declines significantly from 1.99% five years prior to the Event date to 1.56% on the date the board seat is granted to a dissident shareholder. The foregoing results were statistically significant at the 1% level. The mean Q Ratio for the Treatment Group increased from 1.56% on the grant seat date to 1.77% five years post the Event date, reflecting a 2.6% CAGR increase. During the same period, the Control Group experienced a similar increase in its Q Ratio, albeit slightly less than the Treatment Group (See Graph 4). 24 Graph 4 Q Ratio 2.15 Board Seat Grant Date 2.05 1.95 All Target Firms that WON the proxy fight against activist hedge fund: 1.8% CAGR increase in Q Ratio over 5 year period 1.85 1.87 1.77 1.75 1.71 1.65 All Target Firms that granted at least one board seat to a dissident: 2.6% CAGR increase in Q Ratio over 5 year period 1.56 1.55 1.45 1 Event 2 +1 3 +2 4 +3 5 +4 6+5 Years ROIC and ROE reflects a similar pattern to the ROA and Q measures discussed above. Our findings demonstrate a significant decline for both measures during the pre-Event Date period of the activist intervention and a material improvement once a dissident shareholder joined the board of the target firm. Although the Control Group experienced a comparable degradation during the pre-Event Date period – it did not experience a similar increase in improvement compared to the target firms that granted at least one board seat to a shareholder activist. Capital investment policy (CAPEX/Sales) of the target firms for the Treatment Group and for the Control Group experienced a slight decline of fixed asset investment leading into the Event. However, Panel A demonstrates that target firms that granted at least one board to a dissident shareholder continued to experience a decline of CAPEX as a percentage of revenue ex post the Event by approximately 40% of its capital spending prior to the Event. Panel C illustrates that the Control Group increased its capital spending during Event +1 above the Event period. Suggesting that those target firms may have under-invested in fixed assets leading into 25 the Event to manage earnings – but had a “catch-up” period during Event +1 after the incumbent management defeated the dissident in the proxy contest. Notwithstanding the decrease in capital spending as a percent of sales, the target firms that granted at least one board seat to an activist hedge fund experienced significant increases in both ROIC and ROE. In contrast, the Control Group suffered a decline in comparable return measures. A frequently invoked claim by opponents of hedge fund activism is that activist interventions lead to deteriorating operating performance of the target firms ex post the activist event. We empirically tested the operating performance of our data set by comparing target firms in our Treatment Group to the Control Group (See Graph 5). After five years, the Target Firms that granted at least one board seat experienced a slight improvement in operating margin, whereas the target firms that the management team won the proxy contest against the dissident shareholder experienced a -7% CAGR in operating margin. Graph 5 Operating Margin 7.00 6.50 6.00 All Target Firms that granted at least one board seat to a dissident : Slight increase in Operating Margin over 5 year period) Board Seat Grant Date 5.50 5.00 4.81 4.50 4.33 4.35 4.00 All Target Firms that WON the proxy fight against activist hedge fund: 7% CAGR decrease 3.50 3.38 in Operating Margin over 5 year period 3.00 1 Event 2 +1 3 +2 4 +3 5 +4 6+5 Years Our investigation and findings support the alternative hypothesis that hedge fund activism is not detrimental to and does not have an adverse-effect on the long term interests of 26 target firms and their long term shareholders. We find that target firms in our sample dataset underperformed their industry peers on certain operating metrics prior to the activist intervention, which validates the value-oriented characteristics of activist targets. However, those operating metrics progressively improved during the review period subsequent to the board seat grant date. This is further evidence that board representation by activist hedge funds lead to improved operating performance in the long term. In contrast, target firms that won the proxy fight against the activist experienced degradation in certain operating performance metrics compared to industry peers during the review period. 27 b. Event Study Methodology Long-horizon event studies have an extensive history, including the original stock split event study by Fama, Fisher, Jensen, and Roll (1969). As evidence inconsistent with the efficient markets hypothesis started to accumulate in the late seventies and early eighties, interest in longhorizon studies continued. Evidence on the post-earnings announcement effect (Ball and Brown, 1968, and Jones and Litzenberger, 1970), size effect (Banz, 1981), and earnings yield effect (Basu, 1977 and 1983) contributed to skepticism about the CAPM as well as market efficiency. This evidence prompted researchers to develop hypotheses about market inefficiency stemming from investors’ information processing biases (DeBondt and Thaler, 1985 and 1987) and limits to arbitrage (DeLong et al., 1990a and 1990b, and Shleifer and Vishny, 1997). The “anomalies” literature and the attempts to model the anomalies as market inefficiencies has led to a burgeoning field known as behavioral finance. Research in this field formalizes (and tests) the security pricing implications of investors’ information processing biases. Because the behavioral biases might be persistent and arbitrage forces might take a long time to correct the mispricing, a vast body of literature hypothesizes and studies abnormal performance over long horizons of one-to- five years following a wide range of corporate event s. The events might be one-time (unpredictable) phenomena like an initial public offering or a seasoned equity offering, or they may be recurring events such as earnings announcements. Both cumulative abnormal returns (CAR) and buy-and-hold abnormal return (BHAR) methods test the null hypothesis that abnormal performance is equal to zero. The abnormal returns for an event are calculated as follows: t2 CARi (t1 , t2 ) ARit t t1 After calculating event CARs, we then calculate and report the cumulative 28 average abnormal return (CAAR). Where, the CAAR is the arithmetic average of all sample event CARs, and is calculated as follows: 1 CAARi (t1 , t 2 ) N N t2 j 1 t t1 AR it Under each method, the abnormal return measured is the same as the returns to a trading rule which buys sample securities at the beginning of the first period, and holds through the end of the last period. CARs and BHARs correspond to security holder wealth changes around an event. Further, when applied to post-event periods, tests using these measures provide information about market efficiency, since systematically nonzero abnormal returns following an event are inconsistent with efficiency and imply a profitable trading rule (ignoring trading costs). While post-event risk-adjusted performance measurement is crucial in long-horizon tests, actual measurement is not straightforward. Two main methods for assessing and calibrating post-event risk-adjusted performance are used: characteristic-based matching approach (also known as BHAR) and the Jensen’s alpha approach, which is also known as the calendar-time portfolio approach (Fama, 1998 or Mitchell and Stafford, 2000). Analysis and comparison of the methods is detailed below. c. Initial Market Reaction – Cumulative Abnormal Return Table 6 presents our findings of the market reaction to news of that a target firm either granted at least one board seat to a dissident shareholder (Treatment Group) or that the target firm did not grant a board seat (Control Group). We examine the market reaction in the 3-day, 5-day and 21-day windows around the board seat grant date or announcement date via a press release of the results of the proxy contest. This study employs a standard event study methodology and a standard market model to measure normal performance. The market model, which improves on the constant 29 mean return model by controlling for R mt , was calculated as follows: Rit i i Rmt it Where, E ( it ) 0 Var ( it ) i2 The regression coefficients αi and βi are estimated in an ordinary least squares (OLS) regression during the estimation period one year (255 trading days) prior to the event period (event days -305 through -50). The event period consists of 21 trading days centered on the product recall announcement event ( -10 through +10). Three event windows were defined based on the event date, [ -10,10], [-2,+2] and [-1,+1]. As proxy for the return for the market portfolio R mt , both the Center for Research on Security Prices (CRSP) value weighted index and the CRSP equal weighted index was used. Under standard assumptions, OLS is a consistent estimation procedure for the market model parameters. Under the assumption that asset returns are jointly multivariate normal and independently and identically distributed, OLS is also efficient. Panel A presents the market response for Target Firms that granted at least one board seat to a dissident shareholder. The results in Panel A sh ow significant abnormal returns of 1.44%, 1.63% and 3.30% during the 3-day, 5-day and 21-day windows, respectively, surrounding the announcement of board representation. The results in Panel A are consistent with our other findings that investors perceive the value associated with board representation of an activist hedge fund. However, Panel B results indicate a significant negative market reaction of -4.33% during the 21-day window around the announcement that a dissident shareholder did not obtain board representation and the incumbent management team prevailed in a proxy fight against an activist hedge fund. Overall, these results suggest that investors react positively to the announcement that an activist hedge fund will function as a shareholder advocate 30 to monitor management through active board engagement and will operate as a disciplinary mechanism on the target firm. Interestingly, the significant negative market reaction during the 21-day trading window when the incumbent management team wins the proxy fight suggests that certain shareholders (perhaps other hedge funds and arbitrageurs) are not supportive of the status quo. Table 6 Cumulative Average Abnormal Return (CAAR) Around Announcement of Board Engagement Panel A: CAR for Target Firms that Granted at least One Board Seat Market Model (Value Weighted) Window Market Model (Equal-Weighted) N CAAR t-stat p-value (- 1, 1) 409 1.44*** 3.94 0.001 (- 2, 2) 409 1.63*** 3.20 (- 10, 10) 410 3.30*** 3.94 Window N CAAR t-stat p-value (- 1, 1) 409 1.38*** 3.95 0.001 0.001 (- 2, 2) 409 1.60*** 3.26 0.001 0.001 (- 10, 10) 410 3.07*** 3.00 0.001 Panel B: CAR for Target Firms that Management Won Proxy Fight Market Model (Value Weighted) Market Model (Equal-Weighted) Window N CAAR t-stat p-value Window N CAAR t-stat p-value (- 1, 1) 65 0.32 0.30 0.384 (- 1, 1) 65 0.29 0.27 0.936 (- 2, 2) 65 -1.03 -0.95 0.170 (- 2, 2) 65 -0.98 -0.94 0.173 (- 10, 10) 65 -4.33** -1.77 0.039 (- 10, 10) 65 -4.52** -1.89 0.029 *, ** and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively. Given the discernable difference between the market reactions around the announcement of an activist hedge fund obtaining board representation vis-à-vis an incumbent management team winning a proxy fight, we conducted a precision -weighted cumulative average abnormal return (PWCAAR) between the respective groups and tested the significance of the results. Graph 6 illustrates the statistically significant 31 PWCAAR market reaction over a 21-day trading window surrounding the announcement of board representation. Graph 6 Precision Weighted Cumulative Average Abnormal Return (PWCAAR) Around Announcement (period in days) 4.00% 3.00% All Target Firms that granted at least one board seat to an activist hedge fund: 3.15%*** PWCAAR over 21 day Event window 3.15% 2.00% 1.00% 0.00% -10 -9 -8 7 -6 -5 -4 -3 -2 -1 EVENT +1 +2 +3 +4 +5 +6 +7 +8 +9 +10 -1.00% -2.00% -3.00% -4.00% -5.00% -6.00% All Target Firms that Management won the proxy fight: -5.63%** PWCAAR over 21 day Event window -5.63% -7.00% *, ** and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively. 32 d. Long Term Stock Returns We examine long term stock returns for each individual target firm that granted at least one board seat to an activist shareholder. We contrast those results with both Control Groups. To identify whether stock returns are abnormally low or high we use a benchmark for comparative purposes. Such benchmarks of comparison are provided by the Capital Asset Pricing Model (CAPM), the Fama-French three factor model and the Fama-French-Carhart asset-pricing model. These models provide a prediction of the return that “normally” would be expected for a given security during a given period and, therefore, enable us to identifying “abnormal” returns. In particular, using the CAPM, the standard procedure is to estimate an “alpha,” the average excess return that is not explained by co-movement with the market. We estimate the excess return on the market as the value-weight return of all CRSP firms incorporated in the US and listed on the NYSE, AMEX, or NASDAQ that have a CRSP share code of 10 or 11 at the beginning of month t, good shares and price data at the beginning of t, and good return data for t minus the one-month Treasury bill rate from Ibbotson Associates. Specifically, we estimate for each firm i an alpha using the following regression: rit - rf = ai + β1 RMRFt + Eit Similarly, using the Fama-French-three-factor model, the standard procedure is to estimate an “alpha,” the average excess return that is not explained by the three market-wide factors identified in by Fama and French (1993). We estimate for each firm i an alpha using the following regression: rit - rf = ai + βi1 RMRFt + βi2 SMBt + βi3 HMLt + Eit The Fama-French factors are constructed using the six value-weight portfolios formed on size and book-to-market. SMB (Small Minus Big) is the average return on the three small 33 portfolios minus the average return on the three big portfolios. HML (High Minus Low) is the average return on the two value portfolios minus the average return on the two growth portfolios and we estimate the excess return on the market similar to the methodology used in CAPM. Additionally, using the Fama-French-Carhart four-factor asset pricing model, the standard procedure is to estimate an “alpha,” the average excess return that is not explained by the four market-wide factors identified by Fama and French (1993) and by Carhart (1997). We estimate for each firm i an alpha using the following regression: rit - rf = ai + βi1 RMRFt + βi2 SMBt + βi3 HMLt + βi4 M0Mt + Eit We use the same methodology as we used in the Fama-French Asset Pricing Model to estimate the first three factors. We added an additional factor to construct a Momentum factor. The momentum factor is the average return on the two high prior return portfolios minus the average return on the two low prior return portfolios. We use six value-weight portfolios formed on size and prior (2-12) returns to construct the MOM. The portfolios, which are formed daily, are the intersections of two portfolios formed on size (market equity) and three portfolios formed on prior (2-12) return. The daily size breakpoint is the median NYSE market equity. The daily prior (2-12) return breakpoints are the 30th and 70th NYSE percentiles. The six portfolios used to construct MOM each day include NYSE, AMEX, and NASDAQ stocks with prior return data. To be included in a portfolio for day t (formed at the end of day t-1), a stock must have a price for the end of day t-251 and a good return for t-21. For each of the firms in the primarily dataset that granted at least one board seat, we estimate a daily alpha, or abnormal return, for a five year period (annually) prior to date the board seat was granted. In addition, we estimate daily alphas for a five year period (annually) following one day post the board seat grant date. To the extent that firms delist from the sample we 34 incorporate the financial returns up to the delisting date. Additionally, to the extent target firms file for Chapter 11 bankruptcy protection, we incorporate the returns from the firm up to the date of filing for bankruptcy. Panel A in Table 7 provides results with respect to the risk-adjusted excess returns (alphas) we calculated for all target firms that granted at least one board seat to a dissident shareholder. Panel B provides our findings of the risk-adjusted excess returns (alphas) for the Control Group. Table 7 Long Term Risk-Adjusted Excess Returns [Event -5] α (Alpha) Observations Mean Median t-Stat Std. Dev. 442 20.63 *** 10.45 4.64 0.07 α (Alpha) Observations Mean Median t-Stat Std. Dev. 442 6.28 -0.90 1.42 0.07 α (Alpha) Observations Mean Median t-Stat Std. Dev. 442 6.22 0.44 1.42 0.07 Panel A: Target Firms that Granted at least One Board Seat Periods (in Years) [Event -3] [Event -1] [Event +1] [Event +3] Capital Asset Pricing Model (CAPM) 442 442 344 208 -2.42 -1.93 0.77 38.08*** -5.20 -2.75 2.46 36.57 -0.76 -0.88 0.24 3.67 0.09 0.18 0.23 0.20 Three-Factor Asset Pricing Model (Fama-French) 442 442 344 208 -4.75 -0.99 0.21 30.45*** -5.53 -2.56 3.89 29.54 -1.55 -0.46 0.07 2.97 0.09 0.18 0.23 0.20 Four-Factor Asset Pricing Model (Fama-French-Carhart) 442 442 344 208 -5.11* -0.95 -2.61 27.41*** -5.26 -2.62 2.13 27.13 -1.72 -0.46 -0.82 2.68 0.08 0.17 0.23 0.20 *, ** and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively. 35 [Event +5] 128 78.21*** 43.05 4.00 0.18 128 64.75*** 32.86 3.27 0.18 128 63.33*** 31.77 3.20 0.18 α (Alpha) Observations Mean Median t-Stat Std. Dev. α (Alpha) Observations Mean Median t-Stat Std. Dev. α (Alpha) Observations Mean Median t-Stat Std. Dev. Panel B: Management Won Proxy Fight (No Shareholder Activist Board Representation) Periods (in Years) [Event -5] [Event -3] [Event -1] [Event +1] [Event +3] [Event +5] Capital Asset Pricing Model (CAPM) 53 53 53 56 35 24 51.21*** 28.79*** 7.33 5.37 32.40 43.81 36.77 32.29 11.99 4.12 15.92 6.26 3.59 3.19 1.32 0.94 1.05 0.79 0.08 0.09 0.16 0.17 0.24 0.21 Fama-French Asset Pricing Model 53 53 53 56 35 24 34.14** 22.38*** 6.00 3.33 31.02 37.50 24.35 18.66 9.58 4.96 16.39 6.64 2.52 2.59 1.03 0.60 1.03 0.68 0.08 0.08 0.17 0.17 0.24 0.21 Fama-French-Carhart Asset Pricing Model 53 53 53 56 35 24 35.91*** 22.68*** 6.39 1.66 29.16 33.65 26.86 16.80 13.58 3.41 16.01 4.28 2.65 2.59 1.15 0.30 0.98 0.61 0.08 0.08 0.16 0.17 0.23 0.21 *, ** and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively. For each of the periods, we provide both the median and average alpha for all the firms in our sample. We also indicate the statistical significance of our results; however, as is now wellknown in the financial economics literature, the standard error of the average of the estimated alphas understates the unobserved variability in performance, and the reported t-stats should thus be treated as merely suggestive (Fama, 1998). The first row in Table 8 provides our results concerning the risk adjusted excess return (alpha) during each year for the five-year period preceding the board seat grant date. Using the three asset pricing models, we find an alpha during the three-year period ex ante that is negative and economically meaningful. These results, like those obtained with respect to target firm operating performance, are consistent with the view that hedge fund activists target underperforming companies. Additionally, target firms that granted at least one board seat outperformed the Control Group. 36 What is more noteworthy, are the results with respect to the risk adjusted excess return (alpha) during the five-year period following the board seat grant date. The average and the median of the estimated alphas are positive and statistically significant when we use the CAPM model, the Three Factor (Fama-French) Asset Pricing Model and the Four Factor Model. The results provide no support for the negative returns during these periods hypothesized by opponents of activism. More specifically, our results suggest that hedge fund activism generates substantial long-term value for target firms and its long-term shareholders when they act as a disciplinary mechanism to monitor management. Moreover, hedge fund activists that function as disciplinary mechanism to monitor management generate higher risk-adjusted excess returns compared to the Control Group. e. Buy-and-hold abnormal returns (BHAR) Approach In recent years, following the works of Ikenberry, Lakonishok, and Vermaelen (1995), Barber and Lyon (1997), Lyon et al. (1999), the characteristic-based matching approach (or also known as the buy-and-hold abnormal returns, BHAR) has been widely used. Mitchell and Stafford (2000) describe BHAR returns as “the average multiyear return from a strategy of investing in all firms that complete an event and selling at the end of a pre-specified holding period versus a comparable strategy using otherwise similar nonevent firms.” An appealing feature of using BHAR is that buy-and-hold returns better resemble investors’ actual investment experience than periodic (monthly) rebalancing entailed in other approaches to measuring riskadjusted performance. The joint-test problem remains in that any inference on the basis of BHAR hinges on the validity of the assumption that event firms differ from the “otherwise similar nonevent firms” only in that they experience the event. The researcher implicitly assumes 37 an expected return model in which the matched characteristics (e.g., size and book-to- market) perfectly proxy for the expected return on a security. Since corporate events themselves are unlikely to be random occurrences, i.e., they are unlikely to be exogenous with respect to past performance and expected returns, there is a danger that the event and nonevent samples differ systematically in their expected returns notwithstanding the matching on certain firm characteristics. This makes matching on (unobservable) expected returns more difficult, especially in the case of event firms experiencing extreme prior performance. Once a matching firm or portfolio is identified, BHAR calculation is straightforward. A T- month BHAR for event firm i is defined as: BHARi (t, T) = t = 1 to T (1 + Ri,t) - t = 1 to T (1 + RB,t) Where RB is the return on either a non-event firm that is matched to the event firm i, or it is the return on a matched (benchmark) portfolio. If the researcher believes that the Carhart (1997) four- factor model is an adequate description of expected returns, then firm-specific matching might entail identifying a non-event firm that is closest to an event firm on the basis of firm size (i.e., market capitalization of equity), book-to-market ratio, and past one- year return. Alternatively, characteristic portfolio matching would identify the portfolio of all nonevent stocks that share the same quintile ranking on size, book-to-market, and momentum as the event firm (see Daniel, Grinblatt, Titman, and Wermers, 1997, or Lyon, Barber, and Tsai, 1997, for details of benchmark portfolio construction). The return on the matched portfolio is the benchmark portfolio return, RB. For the sample of event firms, the mean BHAR is calculated as the (equal or value-weighted) average of the individual firm BHARs. Additionally, we tested the t CAR / CAR / CAR i i significance of each coefficient using the following equation: or n t BHAR BHARi / BHARi / n 38 We constructed two BHAR portfolios of stocks from 1997-2013 for the Treatment Group and the Control Group. There were over 12,000 observations within the portfolios. We calculated the daily abnormal returns for each target firm relative to the market benchmarks discussed previously. Portfolio I, which is all target firms that granted at least one board seat to an activist hedge fund generated approximately 8 bps/day or 20% annually of risk-adjusted excess return (alpha) relative the market. Consistent with our other findings, our Treatment Group outperformed the Control Group. Moreover, Portfolio I (the Treatment Group) outperformed Portfolio II (the Control Group that incumbent management won the proxy contest) by approximately 5 bps/day or 13% annually. All portfolios had high exposure to small cap stocks and a high value tilt. 39 Table 9 (Portfolio I) Buy-and-Hold Abnormal Return Portfolio of All Activist Board Governed Firms CAPM SUMMARY OUTPUT α βp, RM-RF Coefficients Standard Error 0.08 0.02 0.73 0.01 t Stat 4.55 55.23 P-value <0.0001 <0.0001 Lower 95% 0.04 0.70 Upper 95% 0.11 0.75 THREE FACTOR MODEL (FAMA-FRENCH) SUMMARY OUTPUT α βp, RM-RF βp, SMB βp, HML Coefficients Standard Error 0.06 0.02 0.74 0.01 0.61 0.03 0.43 0.02 t Stat 4.12 60.80 23.73 17.62 P-value <0.0001 <0.0001 <0.0001 <0.0001 Lower 95% 0.03 0.72 0.56 0.38 Upper 95% 0.10 0.76 0.66 0.48 FOUR FACTOR MODEL (FAMA-FRENCH-CARHART) SUMMARY OUTPUT α βp, RM-RF βp, SMB βp, HML βp, MOM Coefficients Standard Error 0.07 0.02 0.72 0.01 0.62 0.03 0.39 0.03 -0.09 0.02 t Stat 4.35 55.80 24.15 15.22 -5.43 P-value <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 Lower 95% 0.04 0.69 0.57 0.34 -0.13 Upper 95% 0.10 0.74 0.67 0.44 -0.06 *, ** and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively. Portfolio II Buy-and-Hold Abnormal Return Portfolio of Control Group CAPM SUMMARY OUTPUT α βp, RM-RF Coefficients Standard Error 0.076 0.026 0.714 0.020 t Stat 2.89 35.71 P-value <0.0001 <0.0001 Lower 95% 0.02 0.68 Upper 95% 0.13 0.75 THREE FACTOR MODEL (FAMA-FRENCH) SUMMARY OUTPUT α βp, RM-RF βp, SMB βp, HML Coefficients Standard Error 0.063 0.026 0.690 0.020 0.401 0.045 0.256 0.041 t Stat 2.44 34.78 8.88 6.26 P-value <0.0001 <0.0001 <0.0001 <0.0001 Lower 95% 0.01 0.65 0.31 0.18 Upper 95% 0.11 0.73 0.49 0.34 FOUR FACTOR MODEL (FAMA-FRENCH-CARHART) SUMMARY OUTPUT α βp, RM-RF βp, SMB βp, HML βp, MOM Coefficients Standard Error 0.065 0.026 0.668 0.022 0.413 0.045 0.236 0.041 -0.069 0.028 t Stat 2.50 30.70 9.10 5.70 -2.49 P-value <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 *, ** and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively. 40 Lower 95% 0.01 0.63 0.32 0.16 -0.12 Upper 95% 0.12 0.71 0.50 0.32 -0.01 f. Calendar-time portfolio approach (Jensen’s alpha) The calendar-time portfolio or Jensen-alpha approach to estimating risk-adjusted abnormal performance is an alternative to the BHAR calculation using a matched-firm approach to risk adjustment. Jaffe (1974) and Mandelker (1974) introduced a calendar time methodology to the financial-economics literature, and it has since been advocated by many, including Fama (1998), Mitchell and Stafford (2000) and Brav and Gompers (1997). The distinguishing feature of the most recent variants of the approach is to calculate calendar-time portfolio returns for firms experiencing an event, and calibrate whether they are abnormal in a multifactor regression. The estimated intercept from the regression of portfolio returns against factor returns is the post-event abnormal performance of the sample of event firms. We implemented the Jensen-alpha approach for a five year period during the pre-event (i.e., prior to the dissident / activist board seat grant date) and then post-event annually for a five year period. In each calendar month over the entire sample period, a portfolio was constructed comprising all firms experiencing the event within the previous month. Since the number of event firms is not uniformly distributed over the sample period, the number of firms included in a portfolio is not constant through time. As a result, some new firms are added each month and some firms exit each month. Accordingly, the portfolios are rebalanced each month and an equal or value-weighted portfolio excess return is calculated. The resulting time series of monthly excess returns is regressed on the CAPM market factor and the three Fama-French (1993) factors as follows: Rpt – Rft = αp + βp (Rmt – Rft) + βp,SMB SMB t + βp,HMLHML t + εpt Where; Rpt is the equal or value-weighted return for calendar month t for the portfolio of event firms that experienced the event within previous T years, Rft is the risk- free rate, Rmt is the return on the CRSP value-weight market portfolio, SMBpt is the difference between the return on the portfolio of “small” stocks and “big” stocks; HMLpt is the difference between the return on the portfolio of “high” and “low” book-to-market stocks; αp is the average monthly abnormal return (Jensen alpha) on the portfolio of event firms over the T41 month post-event period, βp is the beta (the sensitivities) of the event portfolio to the three factors. Inferences about the abnormal performance are on the basis of the estimated αp and its statistical significance. Since αp is the average monthly abnormal performance over the T- month post-event period, it can be used to calculate annualized post-event abnormal performance. Table 8 reports statistics on longterm abnormal returns associated with firms that granted at least one board seat to an activist/dissident shareholder. We report regression estimates and t-statistics from value-weighted calendar-time portfolio regressions. The portfolio holding period, was determined based on actual trading days and indicates the holding period in years relative to the date that the board seat(s) was granted. For example, the portfolio with holding period [Event +1], continually adds target firms that have added an activist/dissident shareholder to their respective board during the year from the date the seat was granted. The portfolio holds these firms until the earlier of December 31, 2013, a delisting date as a result of a Chapter 11 filing or a sale/merger. We report regression results separately for all targets in Panel A of Table 10. αp is the estimate of the regression intercept from the factor model. βp,RM RF is the loading on the market excess return. βp,SMB and βp,HML are the estimates of portfolio factor loadings on the Fama-French size and book-tomarket factors. We obtain the factor returns, market capitalization breakpoints, and monthly risk-free rates from Ken French’s web site at Dartmouth College. 42 Table 10 Calendar-Time Series Regressions (Long-term Abnormal Returns) Panel A: Target Firm Fama-French Calendar Time Series Regressions (N=443) α [Event -3] -0.076 [Event -2] -0.101 [Event -1] -0.050 1.1 1.70* 1.48 Holding Period (in Years) [Event +1] [Event +2] 0.000 0.101 NM 1.26 [Event +3] 0.151 [Event +4] 0.202 [Event +5] 0.252 1.64* 1.35 1.25 βp, RM-RF 0.893 99.53*** 0.864 87.22*** 0.837 65.94*** 0.850 55.50*** 0.851 70.44** 0.852 75.23*** 0.865 79.43*** 0.864 79.16*** βp, SMB 0.750 40.67*** 0.716 35.04*** 0.673 25.50*** 0.668 20.72*** 0.682 26.82*** 0.677 28.36*** 0.671 29.29*** 0.660 28.74*** βp, HML 0.432 24.38*** 70.21% 3881*** 0.440 22.43*** 65.63% 2983*** 0.508 20.15*** 53.73% 1717*** 0.651 21.29*** 47.33% 1254*** 0.514 21.33*** 58.63% 1977*** 0.511 22.61*** 61.71% 2249*** 0.496 22.85*** 64.05% 2485*** 0.503 23.09*** 63.83% 2462*** Adjusted R2 F-Test *, ** and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively. Panel B: Target Firm Fama-French Calendar Time Series Regressions (N=73) α βp, RM-RF βp, SMB βp, HML Adjusted R2 F-Test Holding Period (in Years) [Event +1] [Event +2] 0.076 0.151 0.93 1.19 [Event -3] 0.151 1.3 [Event -2] 0.101 0.86 [Event -1] 0.101 1.33 [Event +3] 0.302 1.35 [Event +4] 0.302 1.23 [Event +5] 0.378 1.23 0.939 43.38*** 0.981 34.85*** 0.933 27.08*** 0.918 30.57*** 0.879 31.43*** 0.843 31.37*** 0.823 30.37*** 0.822 30.06*** 0.759 0.790 0.607 0.602 0.593 0.546 0.528 0.525 18.64 13.63*** 8.24*** 9.48*** 10.36*** 10.13*** 9.83*** 9.69*** 0.46 11.42*** 59.36% 1960*** 0.56 10.80*** 51.43% 1357*** 0.57 8.68*** 35.53% 660*** 0.34 5.41*** 34.39% 589*** 0.34 6.01*** 37.04% 661*** 0.28 5.14*** 36.01% 633*** 0.25 4.64*** 35.30% 614*** 0.27 5.08*** 35.53% 620*** *, ** and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively. During our investigation of the presence of abnormal returns during this period, we employed three standard methods used by financial economists for detecting stock return underperformance. In particular, the study examines whether the returns to targeted firms were systematically lower than what would be expected given standard asset pricing models. Our findings demonstrate that the targeted firms started to underperform relative to the market two-to-three years prior to the board representation. More importantly, we find that those firms generated positive risk-adjusted excess returns (alpha) within the first two-years after the dissident joined the board. Additionally, we find no evidence that target firms 43 experience a “reversal of fortune” during the five-year period following the activist intervention. The long-term underperformance asserted by supporters of the myopic activism claim, and the resulting losses to long-term shareholders due to activist interventions, are not found in the data. CONCLUSION Over the past two decades, hedge fund activism has emerged as new form of corporate governance mechanism that brings about operational, financial and governance reforms to a corporation. Many prominent business executives and legal scholars are convinced that the entire American economy will suffer unless hedge fund activism with its perceived short-termism agenda is significantly restricted. Shareholder activists and their proponents claim they function as a disciplinary mechanism to monitor management and are instrumental in mitigating the agency conflict between managers and shareholders. The vast majority of shareholder activism literature is predicated on Schedule 13D filings. However, we assert that the optimal dataset to empirically test the long-term effects of shareholder activism should be based on board representation of target firms by a shareholder activist. We find statistically meaningful empirical evidence to reject the anecdotal conventional wisdom that hedge fund activism is detrimental to the long term interests of companies and their long term shareholders. Moreover, our findings suggest that hedge fund activism generates substantial long term value for target firms and its long term shareholders when they function as a shareholder advocate to monitor management through active board engagement. Our research fills the important void with respect to the long term efficacy of shareholder activists serving as a disciplinary mechanism on the firm by actively seeking board representation to monitor management. Additionally, we contribute to the literature regarding shareholder activists as self- interested myopic investors at the expense of the long-term interest of the company and its long term shareholders. Moreover, our findings have important policy implications related to the ongoing debate on corporate governance and the rights and roles of shareholders. 44 Although some prominent legal commentators and presiding justices, such as Chief Justice Strine, have called for restrictions on hedge fund activism because of its supposedly short-term orientation, our findings suggest that hedge fund activism generates substantial long term value for target firms and its long term shareholders when they function as a shareholder advocate to monitor management through active board engagement. 45 REFERENCES Ahern, WP. (2009). Sample selection and event study estimation. Journal of Empirical Finance 16 (2009) 466–482. 2009. Ball, R. & Brown, P. (1968). An empirical evaluation of accounting income numbers. Journal of Accounting Research. Volume: 6, Pages: 159–177. 1968. Ball, R. & Brown, P.R. (2013). Ball and Brown (1968): A Retrospective (September 3, 2013). FIRN Research Paper; Accounting Review, Forthcoming. Banz, R.W. (1981). The relationship between return and market value of common stocks. Journal of Financial Economics. Volume: 9, Issue: 1, Pages: 3–18. March 1981, Barber, B., and Lyon, J. (1996), Detecting abnormal operating performance: The empirical power and specification of test statistics, Journal of Financial Economics 41: 359-399. 1996. Barber, B., and Lyon, J. (1997), Detecting long-run abnormal stock returns: The empirical power and specification of test statistics, Journal of Financial Economics 43: 341-372. 1997. Basu, S. (1977). Investment Performance of Common Stocks in Relation to their Price-Earnings Ratio: A Test of the Efficient Market Hypothesis, Journal of Finance, 32, June, 663-682. 1977 Basu, S. (1983). The Relationship between Earnings Yield, Market Value and Return for NYSE Common Stocks: Further Evidence, Journal of Financial Economics. Volume: 12. Pages: 129-156. 1983 Bebchuk, L. (2013). The Myth that insulating boards serves long term value. Columbia Law Review. Vol. 113, No. 6 (OCTOBER 2013) , pp. 1637-1694 Published by: Columbia Law Review Association, Inc. Bebchuk, L.A., Brav, A., & Jiang, W. (2013) The Long-Term Effects of Hedge Fund Activism. Manuscript Draft of July 2013, Comments Welcome. Becht, M., Franks, J., Mayer, C. & Rossi, S. (2009) Returns to Shareholder Activism: Evidence from a Clinical Study of the Hermes UK Focus Fund. The Review of Financial Studies, Volume 22, Issue 8, pp. 3093-3129. 2009. Berle, A. & Means, G. (1932). The Modern Corporation and Private Property. Transaction Publishers. 1932. Black, Andrew; Davies, John; Wright, Philip D. (1998, 2001). In search of shareholder value: managing the drivers of performance. Financial Times Prentice Hall. 2nd ed. ISBN: 9780273650836, 0273650831. 1998 (1 st Ed.). 2001 (2nd Ed.). Boyson, N.M. & Mooradian, R.M. (2011). Corporate governance and hedge fund activism. Review of Derivatives Research. Volume: 14. Issue: 2. Pages: 169-204. July Brav, A. & Gompers, P.A. (1997). Myth or reality? The long-run underperformance of initial public offerings: Evidence from venture and non-venture capital-backed companies, Journal of Finance. Volume: 52. Pages: 1791–1821. 1997. Brav, A., Jiang, W., Partnoy, F., & Thomas, R. (2008). Hedge fund activism, corporate governance, and firm performance. The Journal of Finance. Volume: 63(4). Pages: Brown, L.D., & Caylor, M.L. (2005). A Temporal Analysis of Quarterly Earnings Thresholds: Propensities and Valuation Consequences. The Accounting Review 80 (April): 423-440. 2005. Brown, L.D., & Caylor, M.L. (2006. Corporate Governance and Firm Valuation. Journal of Accounting and Public Policy 25 (July-August): 409-434. 2006. Brown, L.D., & Caylor, M.L. (2009). Corporate Governance and Firm Operating Performance. Review of Quantitative Finance and Accounting 32 (February): 129-144. Brown, S., & Warner, J. (1980), Measuring security price performance, Journal of Financial Economics 8: 205-258. Brown, S., & Warner, J. (1985), Using daily stock returns: The case of event studies, Journal of Financial Economics. Volume: 14. 3-31. Carhart, M.M. (1997). On Persistence in Mutual Fund Performance. The Journal of Finance, Volume: 52, No. 1. (Mar., 1997), Pages: 57-82. 1997 Caylor, M.L. (2010). Strategic Revenue Recognition around Earnings Benchmarks. Journal of Accounting and Public Policy 29 (January-February): 82-95. 2010. 46 Clifford,C.P. (2008). Value creation or destruction? Hedge funds as shareholder activists. Journal of Corporate Finance. Volume: 14. Issue: 4. Pages: 323-336. September Daniel, K., Grinblatt, M., Titman, S., & Wermers, Russ. (1997). Measuring Mutual Fund Performance with CharacteristicBased Benchmarks. The Journal of Finance. Volume: 52, Issue: 3, Pages 1035–1058, July 1997. DeAngelo, H & DeAngelo, L. (1989. Proxy contests and the governance of publicly held corporations, Journal of Financial Economics, Volume 23, Issue 1, June 1989, Pages 29-59, ISSN 0304-405X DeBondt, W.F.M. & Thaler, R.H. (1985). Does the Stock Market Overreact. Journal of Finance. Volume: 40. Pages: 557581. 1985. DeBondt, W.F.M. & Thaler, R.H. (1987). Further Evidence On Investor Overreaction and Stock Market Seasonality. The Journal of Finance. Volume: 42. Issue: 3. Pages 557–581. July 1987. DeLong, J.B., Shleifer, A., Summers, L.H., & Waldmann, R.J. (1990), Positive-Feedback Investment Strategies and Destabilizing Rational Speculation, Journal of Finance. Volume: 45. Issue:2 (June), Pages: 374-397. 1990 DeLong, J.B., Shleifer, A., Summers, L.H., & Waldmann, R.J. (1990). Noise trader risk in financial markets. Journal of Political Economy. Volume: 98, Pages: 703–738. 1 Dodd, P & Warner, J.B. (1983). On corporate governance: A study of proxy contests, Journal of Financial Economics, Volume 11, Issues 1–4, April 1983, Pages 401-438, ISSN 0304-405X Eckbo, B., Maksimovic, V. & Williams, J. (1990). Consistent estimation of cross-sectional models in event studies, Review of Financial Studies 3, 343-365. Eckbo, B., Masulis, R. & Norli, O. (2006). Seasoned public offerings, in B.E. Eckbo, Handbook of Corporate Finance: Empirical Corporate Finance, (Elsevier/North-Holland), Chapter 6. Eckbo, B., Masulis, R. & Norli, O. (2006). Seasoned public offerings: Resolution of the ‘New Issues Puzzle,’” Journal of Financial Economics 56, 251-291. Ertimur, Y., Ferri, F. & Oesch, D. (2013). Shareholder Votes and Proxy Advisors: Evidence from Say on Pay. Journal of Accounting Research, Volume 51, Issue 5, pp. 951-996. December 2013 Fama, E.F. & French, K.R. (1992). The cross-section of expected stock returns. Journal of Finance. 47, 427–465. Fama, E.F. & French, K.R. (1993). Common risk factors in the returns on stocks and bonds. Journal of Financial Economics. 33, 3–56. Fama, E.F. & French, K.R. (1996). Multifactor explanations of asset pricing anomalies. Journal of Finance. 51, 55–84. Fama, E.F. & French, K.R. (1997). Industry costs of equity. Journal of Financial Economics. 43. 153–193. Fama, E.F., (1998). Market efficiency, long-term returns, and behavioral finance. Journal of Financial Economics. 49, 283– 306. Fama, E.F., Fisher, L., Jensen, M.C., & Roll, R. (1969). The Adjustment of Stock Prices to New Information. International Economic Review. Volume: 10, Issue: 1. Gantchev, N. (2013). The costs of shareholder activism: Evidence from a sequential decision model. Journal of Financial Economics, Elsevier, vol. 107(3), pages 610-631. Gerstein, M.D., Haber, S.R., Campion, T., Isaacson S.B., Faris, B.C., Drewry, C.R. (2014). Third Point LLC v. Ruprecht — Activism Confronts the Rights Plan, Latham & Watkins M&A Client Alert, May 2014. Gillan, S.L. & Starks, L.T. (2007). The Evolution of Shareholder Activism in the United States. Journal of Applied Corporate Finance, Volume: 19, Issue: 1, 2007. Grossman, S.J. & Hart, O.D. (1980). Takeover Bids, the Free-Rider Problem, and the Theory of the Corporation. Bell Journal of Economics. Volume: 11 Issue: 1 (Spring) Ikenberry, D., Lakonishok, J. & Theo Vermaelen C. (1995). Market underreaction to open market share repurchases. Journal of Financial Economics. Pages: 181-208. Jaffe, J.F. (1974). Special Information and Insider Trading. The Journal of Business, Volume: 47, Issue:3, Pages: 410-28. 1974. 47 Jensen, M.C. & Meckling, W.H.. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics. Volume 3, Issue 4, October 1976, Pages 305–360 Jones, C.P. and Litzenberger, R.H. (1970). Quarterly earnings reports and intermediate stock price trends. The Journal of Finance. Volume: 25, Issue: 1, pages 143–148, Karpoff, J. (2001). Does shareholder activism work? A survey of empirical findings. Unpublished working paper, University of Washington, Seattle, WA. 2001 Karpoff, J. M., Malatesta, P. H., & Walkling, R. A. (1996). Corporate governance and shareholder initiatives: Empirical evidence. Journal of Financial Economics. Volume: 42. Pages: 365–395. 1996. Klein, A., & Zur, E. (2009). Entrepreneurial shareholder activism: Hedge funds and other private investors. The Journal of Finance. Volume: 64(1). Pages: 187–229. 2009. Kothari, S.P., & Warner, J. (2007), Econometrics of event studies, Chapter 1 in Handbook of Corporate Finance, Volume 1. Loughran, T. & Ritter, J.R. (1995) The new issues puzzle. Journal of Finance. Volume: 50 Pages: 23-51. 1995. Lyon, T.D. (1999). The new wave of suggestibility research: A critique. Cornell Law Review, 84, 1004-1087. 1999. Lyon, J.D., Barber, B.M., & Tsai, C-L. (1999). Improved Methods for Tests of Long-Run Abnormal Stock Returns. The Journal of Finance. Volume: 54. Issue: 1. Mandelker, G, (1974). Risk and return: The case of merging firms, Journal of Financial Economics. Elsevier. Volume: 1(4), Pages: 303-335. 1974. Mitchell, M., & Stafford, E. (2000), Managerial decisions and long-term stock price performance, Journal of Business. Volume: 73: Pages: 287-329. 2000. Mulherin, J.H. & Poulsen, A.B. (1998). Proxy contests and corporate change: implications for shareholder wealth, Journal of Financial Economics, Volume 47. Issue 3. 15 Pound, J. (1988) Proxy Contests and the Efficiency of Shareholder Oversight. Journal of Financial Economics, 1988, Volume 20, Issue 1, pp. 237-265. Romano, R. (1991). The shareholder suit: litigation without foundation? Journal of Law, Economics, and Organization. Volume: 7. Pages: 55–87. 1991. Romano, R. (1993). Public pension fund activism in corporate governance reconsidered. Columbia Law Review. Volume: 93. Pages: 795–853. 1993. Romano, R. (2001). Making institutional investor activism a valuable mechanism of corporate governance. Yale Journal of Regulation. Volume: 18. Pages: 174–251.2001. Seiler, Michael J., (2004). Performing Financial Studies: A Methodological Cookbook, Prentice Hall. First Edition. Shleifer, A. & Vishny, R. W. (1986). Large shareholders and corporate control. The Journal of Political Economy. 94(3) Part 1 461–488.1986. Shleifer, A. & Vishny, R.W. (1997). The Limits of Arbitrage. The Journal of Finance, Volume: 52. Issue: 1. Pages. 35-55. March 1997. Venkiteshwaran, V., Iyer, S.R., and Rao, R.P. (2010). Is Carl Icahn Good for Long-Term Shareholders? A Case Study in Shareholder Activism. Journal of Applied Corporate Finance, Volume 22, Issue 4, pp. 45-57. 2010. Wahal, S. (1996). Pension fund activism and firm performance. Journal of Financial and Quantitative Analysis. Volume: 31(1). Pages: 1–23.1996. Yen, G. & Chen, Y-L. (2005). Proxy Contest, Board Reelection, and Managerial Turnover: Yes, the Proxy Contest Outcome Matters. Managerial and Decision Economics, Vol. 26, No. 1 (Jan. - Feb., 2005), pp. 15-23 Yermack, D. (1996). Higher market valuation of companies with a small board of directors. Journal of Financial Economics. Volume: 40 Pages: 185-211. 1996. 48 Appendix Graph I Long Term Buy-and Hold Returns Portfolio of Target Firms with Activist Board Representation January 1, 1998 – December 31, 2013 450 All Target Firms that granted at least one board seat to a dissident (N=448). 400 350 Economic Recessions Cumulative Returns 300 250 Excess return on the market, value-weight return of all CRSP firms incorporated in the US and listed on the NYSE, AMEX, or NASDAQ. 200 150 100 50 0 S&P 500 Index -50 -100 Years 49 Graph II Long Term Cumulative Market Adjusted Returns (MAR) (Periods in Years) 45.0% All Target Firms that granted at least one board seat to an activist hedge fund: 38.8% Market Adjusted Excess Return after 5 years from the board seat grant date 40.0% 35.0% 38.8% 30.0% 25.0% 21.7% 20.0% 15.0% 10.0% All Target Firms that WON the proxy fight against activist hedge fund: 21.7% Market Adjusted Excess Return after 5 years from the Event date 5.0% 0.0% -5.0% Event +1 +2 +3 +4 +5 -10.0% Years Note: To control for self-selection bias and endogeneity, we constructed a control group (the “Control Group”) of all proxy fights campaigns that did not Appendix result in a board representation during the same period. We used the market-adjusted return (MAR) model, which imposes the following joint restrictions j=0 and j=1. Essentially, it includes market-wide factors but does not account for risk similar to the market model or CAPM. Figure I The Control Group (N=73) is comprised of target firms that were involved in a proxy contest that the target firm incumbent management defeated the Collection Methodology dissident shareholder during the voting process. Therefore,Data we examined not only firms that granted at least one board seat to a dissident shareholder and its ex post effects (the “Treatment Group”), but also companies that were challenged by dissatisfied shareholders and did not suffer the ex post disciplinary effects by an activist hedge fund. 50 Figure 1 Data Collection Methodolgy FactSet S&P Compustat EDGAR Factiva ISS Proxy CapIQ CRSP MergerMetrics All Activist Events 1984-2013 (N=5,063) SharkRepellent All Proxy Contests 1996-2013 (N=1,216) All Activist Events (Non-Proxy Contests) that resulted in Board Representation 1996-2013 (N=418) CONTROL GROUP All Proxy Contests that resulted in Board Representation 1996-2013 (N=621) All Activist Events that resulted in Board Representation 1996-2013 (N=1,039) All Proxy Contests that DID NOT result in Board Representation 1996-2013 (N=595) EXCLUDE SIC 6726 (Mutual Funds, etc), duplicate campaigns by multiple Activists, Bankruptcy data post Filing Date, Missing data (N=300) EXCLUDE SIC 6726 (Mutual Funds, etc), duplicate campaigns by multiple Activists, Bankruptcy data post Filing Date, Missing data (N=212) INCLUDE only activist campaigns by Hedge Funds at Target Firms with Market Caps > $50mm INCLUDE only activist campaigns by Hedge Funds at Target Firms with Market Caps > $50mm Control Group - Final Data Set All Proxy Contests MANAGEMENT WON (Defeated Activist) 1996-2013 (N=73 Firms) Treatment Group - Final Data Set 1996-2013 (N=448 Firms) (N=843 Board Members) All Proxy Contests that did not result in Board Representation 1996-2013 (N=166 Firms) Notes: There is no central database of activist hedge funds. Therefore, we constructed an independent dataset of all activist interventions from 1984-2013 from various sources. Our manually constructed database of shareholder activist events includes 5,063 interventions from 1984-2013. Similar to Gillan and Starks (2007), we define shareholder activist event as a purposeful intervention by “investors who, dissatisfied with some aspect of a company’s management or operations, try to bring about change within the company without a change in control.” Our data collection comprised a multi-step procedure. Our comprehensive dataset of shareholder activist events includes 5,063 interventions from 1984-2013. Of those, 3,899 (77%) filed a 13D. However, approximately 32% of all activist interventions were focused on board engagement, either through a proxy contest (1,216) or dissident campaigns that resulted in board representation via private negotiations (418) with the target management team and board of directors. In our second step, we narrowed our time-frame from 1996-2013 and identified 1,039 activist interventions that resulted in board representation either through a proxy fight or private negotiations. This sample set included 621 proxy fights and 418 activist interventions (non-proxy contests) that resulted in board representation either through a settlement or concessions between the target management and the dissident shareholder. Next, we excluded certain events and if a target firm were to file for bankruptcy protection or liquidation, we included financial information from the target firm up to the Chapter 11 or Chapter 7 filing date. Our final dataset consists of 448 activist interventions (the “Treatment Group”) that resulted in at least one board seat granted to an activist shareholder from 19962013 (see Table 1). A total of 843 board members (see Table 1) were elected at 398 unique target companies. This includes 225 unique dissident shareholders. Of the 448 activist interventions in the Treatment Group, 243 (54%) target firms are still publicly-listed, 186 (42%) were sold/merged and 19 (4%) target firms filed for bankruptcy. By compiling our own database, we avoid some problems associated with survivorship bias, reporting selection bias, and backfill, which are prevalent among other hedge fund databases. To control for self-selection bias and endogeneity, we constructed a Control Group from the set of all proxy fights campaigns that did not result in a board representation during the same period (N=595). Similar to the primary sample set, we excluded certain events for parameter consistency. 51 Table 1 Distribution of Shareholder Activist Board Engagement Campaigns and Dissident Seats Granted All Board Engagements by Hedge Funds at Target Firms with Market Caps over $50mm All Board Engagements 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Total Board Engagements (Completed) 1 2 5 14 5 26 23 25 18 23 74 68 100 90 61 51 69 84 739 Dissident Seats Granted 1 2 16 28 11 54 60 57 38 59 139 128 199 166 102 88 127 179 1454 52 Board Engagements (Completed) 2 4 6 2 6 7 10 10 15 52 52 68 49 33 34 44 54 448 Dissident Seats Granted 2 14 10 3 11 18 17 20 46 102 95 126 86 49 55 77 112 843 Table 5 (Panel A) Return on Assets (ROA) Observations Mean Median t-Stat Std. Error Industry Adjusted ROA Observations Mean Median t-Stat Std. Error Tobin's Q Observations Mean Median t-Stat Std. Error Industry Adjusted Tobin's Q Observations Mean Median t-Stat Std. Error Capex / Sales Observations Mean Median t-Stat Std. Error Operating Margin Observations Mean Median t-Stat Std. Error Return on Invested Capital (ROIC) Observations Mean Median t-Stat Std. Error Return on equity (ROE) Observations Mean Median t-Stat Std. Error Panel A: Target Firms that Granted at least One Board Seat Periods (in Years) [Event -2] [Event -1] [Event] [Event +1] [Event +2] [Event -5] [Event -4] [Event -3] 386 3.09*** 3.46 7.60 0.41 409 2.64*** 3.27 7.17 0.37 416 2.86*** 3.17 9.62 0.30 428 2.83*** 3.20 10.55 0.27 431 2.07*** 2.72 7.13 0.29 270 2.22*** 1.24 3.20 0.69 280 1.19* 0.60 1.74 0.68 284 1.47*** 0.44 2.80 0.52 280 1.36** -0.29 2.20 0.62 274 -0.01 -0.32 -0.02 0.56 360 1.99** 3.17 1.98 0.83 373 1.83*** 1.50 25.08 0.08 390 1.88*** 1.48 29.43 0.06 405 1.82*** 1.46 27.35 0.07 278 0.51*** 0.07 5.85 0.09 287 0.35*** 0.05 4.69 0.07 308 0.24** -0.03 2.52 0.10 [Event -5] [Event -4] 443 8.82*** 3.80 10.84 0.81 [Event +4] [Event +5] 347 0.16 1.12 0.47 0.35 267 0.76* 1.62 1.76 0.43 215 0.95* 1.99 1.78 0.53 178 1.90*** 2.89 3.25 0.59 134 1.64** 3.17 1.98 0.83 281 -0.97* -1.14 -1.80 0.54 220 -2.15*** -1.61 -3.50 0.61 172 -2.12*** -0.67 -3.11 0.68 142 -1.41* -1.12 -1.91 0.74 121 -1.36 0.04 -1.57 0.87 91 -0.69 0.55 -0.59 1.17 410 1.67*** 1.42 29.48 0.06 415 1.56*** 1.29 28.94 0.06 326 1.51*** 1.27 34.23 0.05 252 1.62*** 1.19 28.80 0.05 199 1.72*** 1.27 22.89 0.07 163 1.73*** 1.38 17.44 0.10 113 1.77*** 1.38 17.40 0.10 320 0.19*** -0.01 2.37 0.08 327 0.08*** -0.07 1.28 0.06 332 0.02*** -0.15 -0.34 0.06 257 0.02*** -0.10 0.37 0.06 199 0.09*** -0.05 0.92 0.09 155 0.11*** 0.01 0.91 0.12 127 0.23*** -0.05 1.39 0.16 85 0.18*** -0.09 1.37 0.13 [Event -3] [Event -2] [Event -1] Periods (in Years) [Event] [Event +1] [Event +2] [Event +3] [Event +4] [Event +5] 442 8.48*** 10.88 19.80 0.78 441 7.72*** 10.72 20.80 0.72 437 8.43*** 11.21 21.80 0.75 439 8.49*** 10.89 22.80 0.78 381 2.53** 4.75 2.50 1.01 382 3.68*** 4.05 4.31 0.85 391 2.51*** 4.09 2.95 0.85 448 3.80*** 3.44 7.76 0.49 446 3.39*** 3.74 6.76 0.50 447 3.66*** 3.82 8.81 0.41 446 3.93*** 4.32 10.54 0.37 379 1.61 6.74 0.95 1.70 397 -4.43** 5.36 -1.81 2.45 359 1.38 4.33 1.12 1.23 397 1.00 5.63 0.71 1.41 412 0.47 5.37 0.39 1.21 *, ** and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively. 398 1.35 3.13 1.31 1.03 443 1.10*** 1.72 3.96 0.28 [Event +3] 441 7.80*** 10.83 23.80 0.72 348 6.77*** 9.92 24.80 0.68 267 5.58*** 9.75 25.80 0.57 214 6.78*** 7.55 26.80 0.90 179 4.93*** 7.97 27.80 0.62 136 4.44*** 6.92 28.80 0.64 405 3.77*** 4.33 4.62 0.82 316 2.69** 4.54 2.14 1.26 243 5.91*** 5.13 5.15 1.15 198 4.99*** 4.65 3.01 1.66 165 5.54*** 5.54 3.53 1.57 122 4.18*** 4.35 2.79 1.50 445 3.02*** 3.72 7.84 0.39 446 1.62*** 2.39 3.97 0.41 349 0.20 1.24 0.38 0.54 266 1.18* 2.17 1.82 0.65 215 1.41* 3.16 1.74 0.81 179 2.58** 4.11 2.45 1.05 134 1.61 4.32 0.89 1.81 414 -1.04 4.90 -0.85 1.22 427 -7.29*** 0.90 -4.20 1.73 329 -8.54*** 0.48 -5.25 1.63 252 -7.10*** 1.47 -3.65 1.95 204 -9.24** 2.76 -2.38 3.88 167 -11.74** 4.91 -2.21 5.31 116 -5.03 7.78 -1.02 4.95 53 Table 5 (Panel B) ROA Observations Mean Median t-Stat Std. Error Industry Adjusted ROA Observations Mean Median t-Stat Std. Error Tobin's Q Observations Mean Median t-Stat Std. Error Industry Adjusted Tobin's Q Observations Mean Median t-Stat Std. Error [Event -5] [Event -4] 56 3.61*** 4.23 3.52 1.03 63 4.60*** 4.48 5.16 0.89 65 3.91*** 5.09 2.68 1.46 70 3.56** 5.82 2.04 1.90 71 2.61 4.32 1.28 2.04 72 2.24 3.53 1.44 1.56 55 3.29*** 3.94 3.99 0.83 44 3.59*** 3.38 3.07 1.27 33 3.04 2.80 1.24 2.45 35 1.95 3.02 1.07 1.82 38 0.20 2.39 0.07 2.66 41 4.88 1.92 0.79 6.17 43 5.21 1.91 0.82 6.36 42 6.97 1.15 1.12 6.25 35 14.89** 1.22 2.04 7.28 54 2.14*** 1.76 10.22 0.21 58 2.43*** 1.41 5.30 0.46 59 2.13*** 1.44 8.81 0.24 65 3.19*** 1.55 3.66 0.87 66 1.67*** 1.32 15.08 0.11 67 1.71*** 1.43 15.78 0.11 34 0.77*** 0.33 3.38 0.23 36 1.19* 0.09 1.68 0.71 40 0.38* 0.19 1.57 0.24 47 1.35 0.18 1.19 1.13 47 0.13 -0.03 0.62 0.20 47 0.03 -0.04 0.13 0.21 [Event -4] [Event -3] [Event -2] [Event -1] [Event -5] Capex / Sales Observations Mean Median t-Stat Std. Error Operating Margin Observations Mean Median t-Stat Std. Error Return on Invested Capital (ROIC) Observations Mean Median t-Stat Std. Error Return on equity (ROE) Observations Mean Median t-Stat Std. Error Panel B: Management Won Proxy Fight (No Shareholder Activist Board Representation) Periods (in Years) [Event -3] [Event -2] [Event -1] [Event] [Event +1] [Event +2] [Event +3] [Event +4] [Event +5] 34 3.95** 2.57 2.34 1.70 30 2.95 2.24 1.42 2.08 23 0.41 1.35 0.23 1.75 25 9.06* 0.54 1.66 5.46 17 9.39* 0.19 1.61 5.84 14 6.34 -0.25 1.01 6.25 15 0.71 -1.64 0.22 3.19 52 1.90*** 1.52 9.67 0.20 41 2.02*** 1.59 6.94 0.29 32 2.06*** 1.75 8.35 0.25 27 2.11*** 1.93 6.39 0.33 22 1.87*** 1.57 6.06 0.31 37 0.10 -0.04 0.33 0.32 28 0.18 -0.12 0.40 0.45 22 0.58* 0.10 1.88 0.31 18 0.73* 0.56 1.85 0.39 15 0.41 0.45 1.52 0.27 [Event +2] [Event +3] [Event +4] [Event +5] Periods (in Years) [Event] [Event +1] 72 11.44*** 3.79 2.73 4.19 73 12.32*** 3.81 3.05 4.04 73 8.07*** 4.02 3.25 2.48 72 6.40*** 3.73 5.39 1.19 72 7.50*** 3.54 4.90 1.53 73 6.90*** 3.86 5.28 1.31 55 7.48*** 3.29 3.58 2.09 44 5.55*** 3.03 4.58 1.21 34 4.17*** 2.99 5.25 0.79 29 5.31*** 3.47 4.95 1.07 22 6.34*** 4.12 4.94 1.28 50 0.21 4.23 0.04 4.94 56 1.29 3.04 0.31 4.14 58 6.04 4.75 1.26 4.78 60 7.17* 4.76 4.03 1.78 61 2.02*** 4.06 0.75 2.68 62 8.13*** 4.81 3.08 2.64 47 7.94*** 5.63 2.47 3.21 40 5.54*** 5.15 2.06 2.69 30 9.56*** 6.14 3.47 2.75 27 6.99** 4.43 2.50 2.80 21 5.69* 3.38 1.83 3.10 73 3.51*** 4.09 3.13 1.12 73 5.30*** 4.85 4.85 1.09 73 4.45* 5.17 1.94 2.29 73 3.29 6.36 0.63 5.21 72 6.14*** 5.72 6.93 0.89 72 4.92*** 4.51 5.71 0.86 55 4.75*** 5.61 4.14 1.15 44 5.78*** 6.40 3.35 1.73 34 5.70** 3.21 2.37 2.40 30 3.36 3.12 0.98 3.45 23 -0.16 2.05 -0.05 3.30 54 7.47* 9.62 1.68 4.46 60 9.38* 10.02 1.91 4.92 63 NM 10.66 -0.98 NM 66 NM 9.64 2.05 8.27 68 NM 9.46 1.75 12.79 69 9.76* 6.88 1.68 5.82 53 10.86** 8.82 2.14 5.07 42 4.39 8.75 0.75 5.85 31 1.57 5.82 0.33 4.79 28 -15.60 5.36 -1.17 13.29 23 -23.27 4.28 -1.34 17.33 *, ** and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively. 54 Table 8: Individual Target Firm Regressions (Long-term Abnormal Returns) [Event -5] [Event -4] [Event -3] α (Alpha) Observations 442 442 Mean Median t-Stat Std. Dev. 20.63*** 10.45 4.64 0.07 442 4.90 -2.53 1.30 0.08 -2.42 -5.20 -0.76 0.09 α (Alpha) Observations Mean Median t-Stat Std. Dev. 442 6.28 -0.90 1.42 0.07 442 -3.38 -6.72 -0.92 0.08 442 -4.75 -5.53 -1.55 0.09 α (Alpha) Observations Mean Median t-Stat Std. Dev. 442 6.22 0.44 1.42 0.07 442 -3.79 -7.51 -1.06 0.07 442 -5.11* -5.26 -1.72 0.08 [Event -5] [Event -4] α (Alpha) Panel A: Target Firms that Granted at least One Board Seat Periods (in Years) [Event -2] [Event -1] [Event +1] [Event +2] Capital Asset Pricing Model (CAPM) 442 442 344 263 [Event +3] 208 [Event +4] 173 [Event +5] 128 -6.14** -1.93 0.77 25.37** -7.29 -2.75 2.46 21.05 -2.14 -0.88 0.24 2.07 0.12 0.18 0.23 0.39 Three-Factor Asset Pricing Model (Fama-French) 442 442 344 263 -5.60** -0.99 0.21 20.52* -6.47 -2.56 3.89 18.68 -2.05 -0.46 0.07 1.70 0.11 0.18 0.23 0.39 Four-Factor Asset Pricing Model (Fama-French-Carhart) 38.08*** 36.57 3.67 0.20 56.34*** 38.98 4.35 0.17 78.21*** 43.05 4.00 0.18 208 30.45*** 29.54 2.97 0.20 173 48.39*** 29.42 3.54 0.18 128 64.75*** 32.86 3.27 0.18 442 -5.88** -5.42 -2.21 0.11 208 27.41*** 27.13 2.68 0.20 173 45.60*** 21.06 3.34 0.18 128 63.33*** 31.77 3.20 0.18 442 -0.95 -2.62 -0.46 0.17 344 -2.61 2.13 -0.82 0.23 263 17.79 14.64 1.44 0.40 Panel B: Management Won Proxy Fight (No Shareholder Activist Board Representation) Periods (in Years) [Event -3] [Event -2] [Event -1] [Event +1] [Event +2] [Event +3] Capital Asset Pricing Model (CAPM) Observations Mean Median t-Stat Std. Dev. 53 51.21*** 36.77 3.59 0.08 53 42.31*** 30.62 3.34 0.09 53 28.79*** 32.29 3.19 0.09 53 13.08* 22.93 1.82 0.10 α (Alpha) Observations Mean Median t-Stat Std. Dev. 53 34.14** 24.35 2.52 0.08 53 31.28*** 20.37 2.58 0.09 53 22.38*** 18.66 2.59 0.08 53 10.03 10.79 1.46 0.10 α (Alpha) Observations Mean Median t-Stat Std. Dev. 53 35.91*** 26.86 2.65 0.08 53 31.06*** 20.71 2.64 0.08 53 22.68*** 16.80 2.59 0.08 53 10.42 11.73 1.53 0.10 [Event +4] [Event +5] 53 56 7.33 5.37 11.99 4.12 1.32 0.94 0.16 0.17 Fama-French Asset Pricing Model 45 15.63 5.57 0.82 0.25 35 32.40 15.92 1.05 0.24 31 30.28 9.89 0.70 0.24 24 43.81 6.26 0.79 0.21 53 56 6.00 3.33 9.58 4.96 1.03 0.60 0.17 0.17 Fama-French-Carhart Asset Pricing Model 45 15.72 0.67 0.86 0.24 35 31.02 16.39 1.03 0.24 31 25.69 4.99 0.61 0.23 24 37.50 6.64 0.68 0.21 45 14.33 2.09 0.79 0.24 35 29.16 16.01 0.98 0.23 31 23.02 -0.41 0.54 0.23 24 33.65 4.28 0.61 0.21 *, ** and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively. 53 6.39 13.58 1.15 0.16 55 56 1.66 3.41 0.30 0.17