اینترلاکهای هیئت مدیره و تمایل برای هدف قرار داده شدن در معاملات سهام خصوصی

We examine how board networks affect change-of-control transactions by investigating whether directors’ deal exposure acquired through board service at different companies affect their current firms’ likelihood of being targeted in a private equity-backed, take-private transaction. In our sample of all US publicly traded firms in 2000–2007, we find that companies which have directors with private equity deal exposure gained from interlocking directorships are approximately 42% more likely to receive private equity offers. The magnitude of this effect varies with the influence of directors on their current boards and the quality of these directors’ previous take-private experience, and it is robust to the most likely classes of alternative explanations—endogenous matching between directors and firms and proactive stacking of board composition by management. The analysis shows that board members and their social networks influence which companies become targets in change-of-control transactions.

A large body of literature has examined the monitoring and advisory roles of boards, as well as how board characteristics affect firm value. Recently, a number of papers have extended this work to gauge the ramifications of social relationships among board members. One subset of this work studies social ties between boards and Chief Executive Officers (CEOs), finding that such connections enhance a board's advising ability but possibly at the cost of diminished efficacy in its monitoring function (Kramarz and Thesmar, 2006, Schmidt, 2008 and Hwang and Kim, 2009). Another subset investigates the effectiveness of boards made up of directors who hold multiple board seats. This work suggests that boards with more “interlocked” directors could be poor monitors either because directors’ independence is compromised (Hallock, 1997, Fich and White, 2003 and Larcker et al., 2005) or because board members are simply too busy to keep a watchful eye on management (Fich and Shivdasani, 2006). Investigating the governance implications of directors’ social ties is a natural extension of the corporate governance literature. However, a second perspective on the role of directors’ social networks has received less attention in the corporate finance literature: the board network as a means for information transmission. Sociologists have long viewed each company's board as a node in a firm-to-firm network (overall, the board interlock network) that arises because a large fraction of public company directors are either directors or executives of other firms (Burt, 1983, Palmer, 1983 and Mizruchi, 1992). Individuals who are officers or directors at two or more companies—interlocked directors—become conduits for information, knowledge, and experiences that travel across the active links in the boardroom network. A number of papers have analyzed the impact of the interlock network on financial variables. Davis (1991) examines the diffusion of poison pills in the 1980s, finding that companies with board interlocks to firms that had already adopted the poison pill were more likely to adopt themselves. Khurana (2002) finds that the CEO search process unfolds across the board interlock network as well, as directors consult board-level contacts to identify and vet potential CEO candidates. Cohen, Frazzini, and Malloy (2008) show that mutual fund managers have superior performance on holdings when the investor shares an educational affiliation with a director of the portfolio company, suggesting that membership in an exclusive educational network conveys access to privileged information. Bizjak, Lemmon, and Whitby (2009) trace the spread of options backdating through the board interlock network. Of the work in corporate finance, this paper most resembles ours in its emphasis on the board network as the transmission route for the diffusion of a financial practice. More generally, our paper contributes to the growing stream of research on the effects of social networks in different areas of finance, such as venture capital (Sorenson and Stuart, 2001 and Hochberg et al., 2007), strategic alliances (Robinson and Stuart, 2007 and Lindsey, 2008), and lending markets (Garmaise and Moskowitz, 2003). These works provide empirical support for the idea that social networks are the pipes through which private information flows, and the fact that the agents in (or outside) a network have differential access to this information can influence diverse financial behaviors and outcomes. Our paper investigates the influence of the board network on change-of-control transactions. Specifically, we study the role of board interlocks on a firm's likelihood of being targeted in a private equity (PE)-backed take-private transaction (take private). We capture the spread of PE-relevant experience via the board interlock network by creating a measure of PE Interlocks, which flags director interlocks that occur when a firm has a current director who is interlocked to a past take-private experience through his service as a director or executive of a second company. To illustrate, Eugene Davis was a director of Metals USA in 2005 when it received a take-private offer from a PE firm. Davis also served on the board of Knology Inc. from 2002 to 2007. In years 2006 and 2007 (but not 2002–2005), we treat Knology as having a PE Interlock because it was connected to the Metals USA buyout via Davis. We believe that the presence of a PE Interlocked director on the board (such as Davis on the Knology board after 2006) can increase the likelihood that a company becomes a PE target. We study PE-backed take privates for a few reasons. First is their magnitude. The 473 deals in our sample total to $790 billion in transaction volume, and at the peak of activity in 2007, PE deals made up 45% of all merger and acquisition (M&A) deal value involving public targets. Second, a supportive board facilitates take-private transactions. Although this is also true of M&As, analyzing the take private process is more tractable. Specifically, M&As are strategic transactions in which acquirer–target pairs match through a search process that occurs over a restricted set of firms within which synergies are plausible (Rhodes-Kropf and Robinson, 2008). In contrast, PE deals are often financially oriented transactions in which PE acquirers can (simplistically) be viewed as interchangeable, bringing similar capabilities to the table.1 This allows us to analyze the firm-level hazard rate of going private, instead of modeling matches between specific acquirer–target pairs. We argue that deal experience transmitted through the board interlock network can increase the likelihood that a firm receives a PE offer. A central assumption of our analysis is that prior experience with a private equity deal often favorably disposes a director to future deals, either because it lowers the incremental cost of acquiring deal-relevant information or simply because familiarity with this type of major transaction breeds comfort. In turn, the PE-friendliness of the board matters in the takeover process. Target boards can invoke state-level anti-takeover laws or enact defensive tactics such as poison pills to deter hostile acquirers.2 In addition, as advisers to senior management, directors wield informal power in the take-private process. Given these sources of influence, a PE firm considering a formal offer for a target company is likely to take into account the board's disposition. A favorable board facilitates a quick transaction, whereas an antagonistic one could cause costly and protracted negotiations. Among all US public companies in 2000–2007, we find that firms with one or more directors who have experienced a PE offer at another company are ∼42% more likely to become targets of PE-backed take privates. Also, we show that specific director characteristics and experiences contour the magnitude of the effect of having a PE Interlock. For still-public companies with PE Interlocked directors who had relatively unsuccessful experiences in their past take-private transactions, the PE Interlock effect largely disappears. Likewise, the effect attenuates when the PE Interlocked director has less influence on the board of the still at-risk firm. Thus, PE Interlocked directors’ individual past experiences and influence on their boards affect whether or not boards adopt a pro-PE stance. These results support the interpretation that past experiences are transmitted across the links in the board network. Our findings, like much of the work on social networks, are vulnerable to the question of causal interpretation. Two salient alternative explanations exist for the correlation between PE Interlocks and PE offers. First, the effect could reflect a reverse causal process by which management teams that desire a private equity transaction recruit directors with PE experience to their boards. Prior work indicates that CEOs could influence the board selection process to hire directors who are friendly to their agenda (Shivdasani and Yermack, 1999 and Baker and Gompers, 2003). We refer to this alternative explanation as “board stacking.” Second, directors and firms do not match randomly. Hermalin and Weisbach (2003) and Adams, Hermalin, and Weisbach (2008) argue that because board members are chosen, board characteristics often are endogenously related to firm outcomes. Several authors have modeled board composition as a response to firms’ relative needs for monitoring versus advising (Hermalin and Weisbach, 1998, Adams and Ferreira, 2007 and Harris and Raviv, 2008), and empirical studies have also found associations between board composition and firm characteristics (Boone et al., 2007, Linck et al., 2008 and Coles et al., 2008). If board composition mirrors firm characteristics, the concern for our argument is that the presence of a director on two companies’ boards could reflect an underlying similarity between the two firms, and it could be this commonality that causes each to have an elevated propensity to become a PE target. We refer to this alternative explanation as “director-firm matching.” We conduct an array of supplemental analyses to address these alternatives and to sharpen the identification of a causal effect. We believe that a strength of this paper is the manner in which we exploit the timing of directors’ arrival to and departure from boards, coupled with the timing of the onset of PE Interlocks, to address these endogeneity issues. We show that the evidence is not consistent with board stacking. For instance, we find that PE Interlocked directors typically have many years of tenure on the board of a company when a take-private offer arrives. Likewise, the PE Interlock effect holds specifically for long-seated directors. In both cases, because a multi-year lag exists from when a director joins a company's board and when a transaction is announced, the specific directors who create PE Interlocks are unlikely to have been placed on the board by current management for the purpose of facilitating a private equity deal. Director-firm matching on omitted variables that could be correlated with the likelihood of becoming a PE target is more difficult to definitively exclude, but here, too, we can exploit the sequence of events in the data. First, we show that the PE Interlock effect (on focal firm j) depends on the timing of the onset of the PE experience (at PE target k). Prior to the actual time that a director on company j's board gains PE experience on the board of firm k, company j does not have an elevated risk of being targeted in a takeover. Similarly, reversing this logic, when an interlocked director leaves company k's board prior to the time that firm k attracts a PE offer, we find that the other company j, whose board the director later joins, does not have an elevated risk of being targeted. Both findings suggest that director-firm matching on time-invariant characteristics is unlikely to be driving the core result. We further exploit the timing of interlocks to address the concern that time-varying director-firm matching accounts for the result, and drawing on the literature on the determinants of board composition, we directly control for many of the factors on which directors and firms are likely to match. Overall, the evidence in the paper supports the assertion that the board network influences which firms become PE targets. The remainder of the paper is organized as follows. Section 2 describes the data and summary statistics. Section 3 provides an overview of the methodology and the identification strategy. Section 4 provides baseline results on the drivers of PE offers and examines director-specific interlock effects. Section 5 addresses potential alternative explanations in interpreting the interlock effect, and Section 6 concludes.

The evidence we present shows that board characteristics, including the intercompany network formed by directors who serve on more than one public company board, play an important role in private equity deal generation. We find that still-public firms with board interlocks are much more likely to become targets in PE-backed take-private transactions; that there is an additional, economically significant effect of having interlocked directors who specifically have PE experience through board service at a different company; and that the nature of the individual director's past PE experience and relative influence on a still-public company's board contours the effect of having a PE Interlocked director. We offer a set of explanations for these findings, including that PE-experienced directors are more likely to have ties with key players in the private equity ecosystem, are more likely to be viewed as experts by other members of the board, are more likely to be known by PE firms searching for deals, and could be more likely to advise management to proceed with a deal. All of these factors could influence whether preliminary discussions between PE firms and potential targets are initiated and whether they ultimately rise to the level of a formal offer. We consider two primary alternative explanations: board stacking and director-firm matching on unobserved attributes that correlate with the take-private hazard, which might generate a positive but spurious relation between a firm with a PE Interlocked director and its likelihood of becoming a PE target. Through many supplemental analyses, we find that the PE Interlock effect is robust to these two alternatives. Finally, we believe that the boardroom network is a promising venue in which to further study the influence of social networks on financial market outcomes. First, as our descriptive statistics reveal, this is a relatively dense network. The vast majority of domestic, public companies now have one or more director interlocks with other public firms. Second, board members are central agents in an array of important decisions, including the selection of CEOs, divestitures, M&As, executive compensation, and the adoption of corporate governance practices. At one level, our findings reinforce the idea that a small number of individuals can wield significant influence in company-wide outcomes. At another level, because these types of decision are taken by the entire board, not by solitary directors, our findings ultimately suggest that individual directors are both swayed by and influence others in the network in which they are embedded. Further research into the sources and conduits of boardroom influence can yield important insights that enhance traditional understandings of the drivers of corporate financial behavior.