Abstract
Building on the institutional perspective on capital markets, we examine the process of legitimation that underpins investor valuation of initial public offerings in the context of institutional polycentricism. We focus on the impact of board interlocks of the CEO and internal and external board members on investor perceptions of initial public offering firms in the United States and United Kingdom. We find that the extent of board members’ interlocks relates positively to the extent of the CEO’s interlocks, but this relationship is stronger in the United Kingdom than in the United States. More extensive interlocks lead to higher valuations in the United Kingdom than in the United States. This is the result of differences in institutional confluences that underpin corporate governance in the two culturally related countries.
An initial public offering (IPO) is a critical stage in the life of a firm (Bruton, Chahine, & Filatotchev, 2009). To improve investors’ perceptions of an IPO’s value in the highly uncertain environment of new listings, firms seek to show that they meet investors’ expectations, including corporate governance expectations (Certo, 2003; Sanders & Boivie, 2004). The governance mechanisms a firm adopts hinge on not only regulation and economic efficiency concerns but also perceptions of the firm’s legitimacy in the capital markets where that IPO takes place (Carruthers & Kim, 2011). Such perceptions form a “socially constructed historical pattern of material practice, assumptions, values, beliefs, and rules by which individuals reproduce their material subsistence . . . and provide meaning to life” (Thornton & Ocasio, 1999: 804). Here we examine investor responses to the CEO and other board members’ external directorships, or board interlocks, since prior studies consider such external linkages to be one of the most common drivers of organizational legitimacy (Certo; Deephouse & Suchman, 2008; Filatotchev & Bishop, 2002; Higgins & Gulati, 2006). Unlike previous IPO board studies grounded in agency perspective, we focus on institutional processes that link external board memberships of the IPO firm’s CEO and other board members and how these external connections affect investors’ perceptions of the value of IPOs. We then go further and examine the impact of the differences in multiple institutional environments on the development of IPO board interlocks and valuation outcomes. Specifically, we ask the following question: How do institutional differences affect both the formation of board interlocks and their impact on the IPO firm’s valuation? Our research focus is relevant to IPO studies since it provides a more contextualized, institutionally embedded perspective on governance practices and their organizational outcomes in an environment of high uncertainty (Bell, Filatotchev, & Aguilera, 2014).
A growing body of research focuses on the organizational outcomes of the CEO and other board members’ external ties in IPO firms, but the results so far are ambiguous. For example, studies grounded in the resource dependency perspective suggest that board interlocks enhance a firm’s access to external expertise and critical resources (e.g., Certo, 2003; Chahine, Filatotchev, & Zahra, 2011; Chen, Hambrick, & Pollock, 2008; Pollock, Chen, Jackson, & Hambrick, 2010). Higgins and Gulati (2003), for example, show that IPO firms where directors have affiliations with prominent organizations are more likely to attract a prestigious investment bank–underwriter. However, employing an agency perspective, Filatotchev and Bishop (2002) found that too many external ties may reduce the board members’ monitoring capacity and undermine IPO stock market performance. Thus, by focusing either on just operational or on distributive efficiency aspects of governance, existing research perspectives offer one-sided, and often conflicting, views on the role of board interlocks.
A sociological perspective on financial market behavior suggests an alternative theoretical approach to IPO governance by arguing that stock market reactions to firm-level governance factors are socially constructed (Bell et al., 2014; Carruthers & Kim, 2011; Zajac & Westphal, 1995, 2004; Zuckerman, 1999). From this perspective, stock market valuations of IPOs are an outcome of investors’ perceptions of an IPO firm’s legitimacy rather than rational, efficiency-centered investor decisions. Legitimacy is the “generalized perception or assumption that the actions of an entity are desirable, proper, or appropriate, within some socially constructed system of norms, values, beliefs, and definitions” (Suchman, 1995: 574). Thus, in the setting of uncertainty associated with the process of an IPO, investors tend to focus on institutionalized rules when evaluating the quality of IPO firms (Pollock et al., 2010). These rules are formed by macroinstitutions that frame the process of investor assessment of the firm (Batjargal, Hitt, Tsui, Arregle, Webb, & Miller, 2013; Holmes, Miller, Hitt, & Salmador, 2013). It remains unclear, however, what institutional mechanisms drive internal IPO firms’ responses to institutional pressures, in terms of recruiting a CEO and other board members with extensive interlocks, and how these external connections of board members ultimately affect investors’ perceptions of IPO firms’ value.
This brings us to another important theoretical gap in the IPO governance research—the tendency to focus on a single nation, particularly the United States, assuming a degree of universality in the resulting theoretical and empirical findings. Researchers have increasingly recognized that the institutional setting has significant implications for governance theories and their empirical tests (Bruton, Ahlstrom, & Li, 2010). Therefore, we explore the potential differences in effects of the board member interlocks (internal board members, including specifically the CEO, and external board members) on investor perceptions in the United States and the United Kingdom. The “law and economics” research (La Porta, Lopez-de-Silanes, Shleifer, & Vishny, 1998) considers these two countries as typical representatives of a “common law family” of corporate governance in which firm-level governance factors should have similar salience. We extend prior research and argue that even when nations share a core approach to shareholder value maximization, they still may have significant differences in their institutional features that have a profound impact on the IPO firm’s board development process and associated investor valuations.
We make a number of contributions here to the literature. First, we extend prior research focused on firm-level responses to the legitimation pressures (e.g., Pollock, Rindova, & Maggitti, 2008; Shipilov, Greve, & Rowley, 2010) to the context of external connections of IPO boards. By exploring how board interlocks of the firm’s independent and executive board members relate to its CEO’s external directorships, we provide insight into mechanisms that facilitate (or impede) the adoption of an institutionally accepted practice.
Second, we investigate how institutional factors affect the value placed on board connections in the eyes of potential investors in an IPO firm in the United States and the United Kingdom. The ability to examine two settings that share many commonalities allows a contrast that theoretically expands the understanding of institutional theory of corporate governance. More specifically, we build on the notion of institutional multiplicity, or the confluence of different types of interrelated institutions, including regulatory, normative, and sociocognitive institutions (Batjargal et al., 2013; Holmes et al., 2013). Our theorizing suggests that differences in institutional confluences may cause differences not only in the patterned association between board interlocks of the CEO and other directors but also in investor responses to these governance characteristics of IPO firms. Overall, we contribute to corporate governance theory by suggesting that institutional multiplicity affects the formation of IPO firms’ governance characteristics and their organizational outcomes, such as investor valuations.
Third, our research contributes to the debate on board members’ being “stretched” by several directorships (Beasley, 1996; Core, Holthausen, & Larcker, 1999) and to the argument that internal, nonfinancial costs accompany well-connected board members since CEOs and other insiders may perceive adding a prestigious new director as a threat to their power or standing in the firm’s status order (Acharya & Pollock, 2013). In this regard, our study is relevant not only to academics but also to practitioners since it provides insight into whether interlocks of directorships, depending on their institutional setting, add value or detract from value creation. This evidence base may help to prevent distortions in the allocation of resources to jobs that reduce allocative efficiency of the director labor market (Westphal, 2010).
Finally, we make a number of empirical contributions to IPO research. To explore our theoretical predictions, we gathered a unique sample of matched IPOs in the United States and the United Kingdom, which controls for sample bias associated with multicountry IPO studies. We also deploy research methods that mitigate potential endogeneity problems associated with different types of board interlocks, which represents a methodological advance compared to previous IPO studies.
Theoretical Foundation and Hypotheses
IPO firms are organizations that offer their shares to the public market as they are moving from private to public ownership. Part of the effort to present its stock to the public requires the top management of the IPO firm to convince the investment community that the firm has met its expectations. This process aims at increasing the legitimacy of the firm in a setting where analysts make valuation decisions under conditions of high uncertainty (Pollock et al., 2008; Zajac & Westphal, 1995, 2004). In this setting, the firm’s management must not only meet the legal regulatory guidelines but also conform to “the normative expectations of powerful constituents for organizational behaviour” (Westphal, 2010: 319). In the IPO context, these constituencies may include the investment banks, financial analysts, and public market investors who evaluate what the composition and nature of the board should be if the IPO is to succeed (Certo, 2003; Chahine et al., 2011). As Shipilov et al. argue, “The logic of board reform is a socially constructed pattern of practices, assumptions and rules” (2010: 848). As a result, the assumptions and values of the stock market participants that shape the actions of firms will help direct the formation of the IPO firm’s corporate governance.
One widely shared belief in the context of corporate boards is that board member interconnections with other actors will positively affect the performance of the IPO firm (Filatotchev & Bishop, 2002). For example, board members with extensive external interlocks can contribute their superior human and social capital to enhance the substantive functioning of the firm by providing access to information, strategic and administrative expertise, and external resources that an IPO firm desperately needs (Certo, 2003; Chahine et al., 2011; Chen et al., 2008; Pollock et al., 2010). Board members with external interlocks also help to overcome a potential “legitimacy deficit” that IPO firms suffer in the eyes of public market investors since “prestigious or legitimate persons or organizations represented on the focal organization’s board provide confirmation to the rest of the world of the value and worth of the organizations” (Pfeffer & Salancik, 1978: 145). In addition, directors with extensive external board membership are careful to preserve their standing in a social order, so they will act to increase a firm’s success through mutual monitoring. As Acharya and Pollock summarize these arguments, “The central premise of this research stream is that firms in ambiguous circumstances, particularly young companies that lack established track records or prior performance histories, seek to reduce others’ uncertainties by forging exchange relationships with prestigious or high-status actors” (2013: 1396). Views expressed by stock market practitioners support these arguments. For example, B. R. Evans, managing partner in Summit Partners in Boston, provides the following advice to companies and investors: “Look for board experience. People who have served on boards can hit the ground running, because they already understand the critical issues of audit, compliance, finance, and strategy, as well as other dynamics that can affect board performance” (2007: 2). 1
We argue that the institutional perspective offers an important theoretical mechanism that links board interlocks with investor perceptions of the IPO firm’s value. The sociological studies of financial markets point out that stock market reactions to various board characteristics are socially constructed and not necessarily reflective of short-term benefits of the wealth distribution efficiency in the context of agency theory. Within this stream of research, directorship interlocks are one of the most common legitimacy measures (Certo, 2003; Deephouse & Suchman, 2008; Higgins & Gulati, 2006). For example, Bitektine suggests that director interlocks provide “linkage legitimacy” (2011: 156), or legitimacy based on an organization’s linkages with legitimate social actors in its environment. Following Deephouse and Suchman (60), one can argue that IPO board interlocks signal to investors three measures of legitimacy: pragmatic (directors’ competence), moral (directors’ propriety through acceptance by other companies), and cognitive (directors’ conventionality, since the directors’ linkages to other organizations help external assessors perceive the firm as acting—or just existing—in ways that are comprehensible and recognizable). Therefore, in a highly uncertain environment associated with an IPO governance, board characteristics are a product not only of coordinative demands imposed by the market efficiency concerns but also of norms legitimizing the adoption of appropriate governance practices (Zajac & Westphal, 2004). This perspective of financial markets focuses less attention on the individual efficiency outcomes of board interlocks that form the core of previous IPO research and more on theoretical efforts to understand how governance mechanisms affect a firm’s legitimacy through perceptions of external assessors of organizational legitimacy or the stock market “audience” (Deephouse & Suchman).
Here we employ the institutional perspective to understand the role of board interlocks in an IPO firm. We focus specifically on IPO valuations by investors since such valuations reflect the investors’ perceptions about the firm rather than a rationally and objectively calculated monetary value (Bell et al., 2014). IPO firms may respond to the powerful institutional pressures in their environment by adopting governance features that match investor perceptions. Our view here of interlocks is in contrast with prior IPO studies that generally take board interlocks as given, exogenous variables that affect IPO stock market performance. By adopting an institutional lens, we can explore more nuanced, institutionally embedded mechanisms that determine the IPO firm’s response to external legitimacy pressures and decisions to engage in activities that have potential benefits in terms of IPO valuations. Our emphasis here is on the assumption that constellations of interlocking directorships in an IPO firm are far from being randomly distributed; they are subject to institutional processes that link CEO and other board members’ external connections. More specifically, we argue that, to build the IPO firm’s legitimacy, the well-connected CEOs will seek out board members with strong board interlocks that reflect their strong connections with others outside of the firm. In other words, the extent of the CEO’s interlocks affects the internal and external board member interlocks that contribute to the overall legitimation outcome for a given IPO firm.
We extend these arguments further by building on research about the confluence of macroinstitutional factors associated with regulatory, normative, and sociocognitive institutions in different national settings (Batjargal et al., 2013; Bell et al., 2014; Holmes et al., 2013). We argue that the IPO board development mechanism and its organizational outcomes are not universal; instead, they depend on the specific confluence of institutional characteristics of stock markets within which the IPO firm operates. The United States and the United Kingdom share many cultural and institutional characteristics, such as a common law foundation (La Porta et al., 1998). However, each nation has a specific pattern of multiple institutional confluences leading to different approaches to corporate governance, and evidence indicates that this difference affects investors’ perceptions of firm-level characteristics of IPO firms (Moore, Bell, Filatotchev, & Rasheed, 2012). Therefore, here, we also explore the impact of institutional confluences on the interrelationship between governance characteristics and valuation of the IPO firm.
Boards Directors and Their Interlocks
Our previous arguments suggest that scholars should consider board directors’ interlocks as a legitimation driver aimed at meeting investors’ expectations about the quality of the firm’s governance. What is not clear within this legitimation framework is how the firm responds to the institutional pressures to develop a more legitimate board, a key question within the institutional perspective on corporate governance of the firm (Acharya & Pollock, 2013). Prior evidence suggests that the IPO firm’s selection of board members (both internal and external) is not exogenous and depends upon the top executives’ characteristics (Filatotchev & Bishop, 2002). Thus, in looking deeper at the various types of board members, we examine external and internal board members’ interlocks in light of their relationship to the CEO’s interlocks. Such an approach is consistent with the work of Shipilov et al. (2010) and Pollock et al. (2008), who have developed models of adoption and expansion of legitimation practices based on the social mechanisms of interrelationships within the firm’s board.
Prior IPO studies emphasize the importance of CEO characteristics for the ultimate success of a firm’s listing (Bruton et al., 2009; Certo, Covin, Daily, & Dalton, 2001; Chahine et al., 2011). Entrepreneurship research also strongly argues that the CEO of the firm brings a rich range of resources to the firm, in particular through his or her external connections (Bruton et al.). Therefore, by appointing a well-connected CEO, the firm indicates its compliance with the objectives defined by institutional pressures underlying the practice (Shipilov et al., 2010). If this supposition is correct, then how do IPO firms respond to the external pressure to gain legitimacy by appointing other, non-CEO members of its board?
Acharya and Pollock (2013) suggest that recruiting other prestigious directors is a complex, multifaceted process that affects, and is affected by, the CEO. Prior research on director selection indicates that CEOs tend to favor director candidates who have similar backgrounds, experiences, and credentials (Westphal, 2010). Therefore, CEOs with extensive external corporate connections will more likely associate with directors who have extensive external interlocks as a result of homophily, or the desire to associate with others who share similar social standing, values, and beliefs (Acharya & Pollock). This steam of research also suggests that a primary motivation to accept a board seat by a non-CEO director is establishing and/or maintaining membership in the corporate elite. Firms with preexisting “stores of prestige” have an easier time recruiting other prestigious actors; hence, firms headed by well-connected CEOs will likely attract outside directors with extensive interlocks. Of course, one can also argue that insider directors are recruited by pre-IPO firms mainly for their skills. However, prior studies show that being involved in other boards is part of the director’s skill set the focal firm values since board connections provide important strategic expertise gained from experiences in other firms (Pollock et al., 2010). As Acharya and Pollock summarize, “All else being equal, prestigious CEOs are not only likely to enhance their own prestige by affiliating with prestigious directors; they will also have an easier time recruiting prestigious directors because of the status affiliation benefits they offer” (1400).
These arguments suggest that IPO firms may respond differently to legitimacy pressures depending on characteristics of their CEOs. The theoretical mechanism outlined above frames the linkage between the external connections of the IPO’s CEO and those of other board members. When the IPO firm adopts legitimization actions through the hiring of a well-connected CEO, it sets into motion similar practices with regard to recruiting other directors with extensive interlocks in a manner that institutional theorists describe as “waves of legitimation” (Shipilov et al., 2010: 847), “information cascades” (Pollock et al., 2008: 336), or “prestige snowball effect” (Pollock et al., 2010: 11). The studies using these various terms focus on different organizational contexts, yet there is commonality in that the underlying theoretical mechanisms are based on a suggestion that “organizations that have adopted logic-defining practices continue adopting logic-extending practices” (Shipilov et al.: 847).
Therefore, we argue that an IPO firm, after appointing a CEO with extensive board interlocks, will likely appoint other executives and independent directors with similarly extensive board interlocks. The CEO’s perception of the importance of board members’ social capital from their interlocks to IPO success and the strong desire for legitimacy of the IPO firm encourages firms with highly interconnected CEOs to attract highly interconnected individuals, from either inside or outside the firm, to be on the board of directors. Therefore, we hypothesize that:
Hypothesis 1: In the IPO firm, the extent of external and internal board members’ interlocks is positively related to the extent of the CEO’s interlocks.
Board Legitimacy and Different Institutional Confluences
Recent institutional research on corporate governance indicates that investors’ perceptions of various firm-level governance characteristics are embedded in macroinstitutional contexts within which investors make legitimacy evaluations (Bell et al., 2014; Moore et al., 2012). Institutional theorists argue that investors make legitimacy judgments in “action spaces” where individual entities interact socially. However, overlapping external factors, such as rules-in-use, attributes of the community, and physical conditions (e.g., physical distance between the actors), affect this interaction (Ostrom, 2010). As Ostrom argues, “The set of external variables impacts an action situation to generate patterns of interactions and outcomes that are evaluated by participants in the action situation” (2010: 647).
A number of institutional theorists have developed these ideas further by suggesting the characteristics of institutional environments include multiplicity, which scholars have defined as the confluence of different types of interrelated institutions (Batjargal et al., 2013; Holmes et al., 2013). As Batjargal et al. argue,
The confluence of multiple institutions is theorized to have qualitatively different effects on outcomes than a single institution or several institutions, because the confluence is characterized by dynamic interaction, mutual reinforcement, and a co-integrated and non-separable nature of diverse institutional rules and norms within the entire institutional order. (2013: 1025)
These authors focus on the confluence of regulatory, normative and sociocognitive macroinstitutions or “first-order” boundary conditions (Kraatz & Block, 2008), which, in turn, hinge upon the cultural and historic developments in particular countries (see Holmes et al. for a detailed discussion).
The “law and economics” perspective considers the U.S. and U.K. economies as members of the same family of “common law” tradition, which shares many common characteristics associated with capital markets (La Porta et al., 1998). Indeed, the fundamental systems of corporate governance in the United States and the United Kingdom share a common premise that the goal of corporate governance is to provide protection of shareholders in general and institutional investors in particular (Zajac & Westphal, 2004). However, despite sharing this common core objective, these two systems have key differences in terms of the confluence patterns among formal and informal institutional parameters (Moore et al., 2012), which may have considerable impact on the mechanism we describe in Hypothesis 1.
In terms of the rules-in-use, or regulatory institutions that affect the firm-level board development process, the United States emphasizes formal regulation, while the United Kingdom emphasizes informal voluntary codes. In recent years, the United Kingdom has incorporated this tradition into its Corporate Governance Code, which commonly relies on a self-enforcement of governance rules underpinned by a “comply-or-explain” system of regulation. At present, only one provision in the U.K. Code imposes a numerical restriction on board interlocks, according to which the board should not agree to a full-time executive director taking on more than one nonexecutive directorship in a Financial Times Stock Exchange (FTSE) 100 company or the chairpersonship of such a company. However, this provision focuses on board interlocks of directors in the largest U.K. companies, and it places no restrictions on external board memberships of non-CEO directors. The introduction of more restrictive rules would be unpopular in the United Kingdom, especially as investors generally view multiple directorships as less problematic in the United Kingdom than in other European countries (Barker, 2011). The voluntary nature of compliance in the United Kingdom means that extensive board interlocks do not undermine the IPO firm’s regulatory legitimacy as long as investors value benefits associated with directors’ external ties.
Moore et al. (2012) indicate that the voluntary regulative approach found in the United Kingdom stands in contrast to the more formal regulative traditions found in the United States. The United States has an extensive body of securities and corporate law at both the federal and state levels. The most recent example of the hard law approach found in the United States was the passage of the Sarbanes-Oxley Act (SOX) in 2002, in which the U.S. Congress mandated new and more stringent governance regulations and increased the costs of noncompliance by all public firms (Keenan, 2004). Through legislative action, SOX required firms to put in place a number of measures intended to enhance the overall board efficiency, including a requirement for boards to disclose interlocks in proxy statements. The SOX is just one legislative effort that mandated governance and heightened transparency of public firms in the United States. For example, Section 8 of the Clayton Act forbids any person from simultaneously serving as a director or officer in any two corporations (excluding banks, banking associations, and trust companies) that by virtue of their business and location of operation are competitors. This law effectively limits a scope of within-industry board interlocks in the United States. Therefore, U.S. investors focus more on regulatory constraints on the development of board interlocks than do U.K. investors.
In terms of normative and sociocognitive institutions, the U.S. stock market participants come from across the nation. In contrast, the stock market participants in the United Kingdom tend to engage in dense formal and informal networks within the City of London, which some authors describe as a unique model of “gentlemanly capitalism” (Cain & Hopkins, 1986; Moore et al., 2012). The United Kingdom has a long history of “upper class” elitism combined with a close physical proximity of social actors and the resulting dense interconnections among actors, where an individual’s reputation is critical to doing business and informal, socially embedded rules can effectively govern firm behavior (Cain & Hopkins). A large number of guilds and “worshipful companies” within London’s Square Mile provide a platform for various kinds of networking. These normative institutional aspects are reinforced by the United Kingdom’s system of professional and investor associations, such as the Institute of Directors, the Association of British Insurers (AIB), and the National Association of Pension Funds (NAPF), that facilitate informal contacts between investors, companies, and board directors. A key symbol of these dense interconnections among actors is the proliferation of board linkages among organizations that scholars have recognized (Filatotchev & Bishop, 2002).
Burt (1987) suggests that field-level institutional practices create pressures toward practice adoption even in the absence of direct interfirm ties. Therefore, the dense network-centered business environment in the City of London provides a strong institutional pressure for the firm-level development of boards with extensive external linkages that go beyond personal interlocks of the CEO. This environment stands in contrast to the U.S. capital markets, where “action spaces” are characterized by significant spatial dimensions and reliance on the transactional rules rather than relationships grounded in a history of class elitism, as in the United Kingdom. Therefore, these differences in the “action space” boundary conditions put lesser field-level institutional pressures on U.S. IPOs in terms of board interlock practice adoption.
Finally, the United States and United Kingdom have significant differences in terms of cognitive institutional aspects associated with CEO power. Powerful CEOs are more likely to perceive adding a prestigious new director as a threat to their standing in the firm’s status order. As Acharya and Pollock argue, “While directors in general represent the corporate elite, prestigious directors represent this elite group’s inner circle” (2013: 1398). Therefore, they may have enough status and ambition to exert social influence that a powerful CEO will resist. They may even engage in “negative impression management” by talking down the roles of other board members who represent a threat to their status (Westphal, 2010). However, the CEO role differs significantly—especially in terms of power—in the two countries. Keenan (2004) provides evidence that in the United States, the chairperson and CEO are the same person in a majority of the Standard & Poor’s 500 companies, and this person often also has the title of the company’s “president.” However, this combining of roles is rare in the United Kingdom. In addition, Filatotchev and Dotsenko (2015) indicate that companies in the United Kingdom often appoint a senior independent director whose responsibility is to serve as a communication channel between shareholders and the board, which limits further the CEO’s power. Overall, limiting the CEO’s power is a key approach adopted by the U.K. institutional investors and their influential professional associations, such as the AIB and the NAPF. 2 Therefore, bearing in mind these significant institutional differences associated with the extent of CEO power, U.K. companies are more likely to appoint well-connected board members than are their U.S. counterparts.
To summarize, we expect the development of IPO board interlocks to be different in the United States and the United Kingdom as a result of the unique confluences of formal and informal institutions in each country. In the United Kingdom, where informal relationships dominate the stock markets, the physical proximity of actors and historically powerful social elites encourage the development of dense social interconnections among actors, and where resistance to changes in status ordering is relatively low, one could expect to find a strong common belief among company leaders and stock market “audiences” that board members need to strengthen interconnections through interlocks in order to achieve the firm’s legitimacy. As a result, U.K. IPO firms will more likely seek out both internal and external board members who have the greatest interlocks possible. This situation does not mean that investors do not appreciate interlocks in an institutional setting like the United States. However, it is a difference in magnitude. Therefore, we expect that the process of a firm’s compliance with institutional pressures will differ in the United Kingdom and the United States, with the internal response being more prominent in U.K. IPOs compared to their U.S. peers. Board interlocks in the United Kingdom contribute to normative legitimacy, and the voluntary nature of compliance to the U.K. Corporate Governance Code does not undermine the firm’s regulatory legitimacy. In the United States, however, a shift of emphasis on the board efficiency, CEO resistance, and regulatory constraints may limit what the IPO firm aims to achieve through extensive board interlocks. Therefore, we hypothesize that:
Hypothesis 2: Compared to the United States, the extent of the IPO’s internal and external board members’ interlocks is more positively related to the extent of the CEO’s interlocks in the United Kingdom.
Board Directors’ Interlocks and IPO Valuation
The institutional perspective on financial markets outlined above suggests that “stock market reactions are based on investor perceptions about the sources of efficiency rather than on any unchanged standard of efficiency” (Zajac & Westphal, 2004: 450). Therefore, by relying on extensive directors’ interlocks, an IPO firm may gain a higher level of legitimacy among investors and positively affect their perceptions of its “true” value in the environment of high uncertainty associated with new listings.
Above, we argue that the process of legitimation encourages an IPO firm’s belief in the value of its board members’ interlocks. We develop this argument further and suggest that the stock market should view this practice adoption positively. Within the context of firms that are making their capital market debut through an IPO, the “perceptions of organizational legitimacy shape investor behavior” (Tost, 2011: 686). Board interlocks in an IPO bring cognitive legitimacy (Scott, 1987) because the interlock is “understood, recognizable, and located within the set of the widely held cognitive structures of its institutional environment” (Sanders & Tuschke, 2007: 33). From the normative legitimacy perspective, investors’ perceptions of IPO value will improve because outside investors recognize the ability of the CEO and outside directors to rely on their interconnections with other actors to secure customers, suppliers, and funding for the new venture. As a result, both the interlocks of the CEO and the interlocks of other board members affect IPO investors’ valuations (Chahine et al., 2011).
However, scholars should not assume the universality of the positive impact of board interlocks. As in Hypothesis 2, subtle differences in the macroinstitutional confluences outlined above may generate different impacts from the firm’s adherence to institutional pressures on investors’ perceptions of an IPO. Zajac and Westphal argue that “investor perceptions of the value of corporate policies are influenced by prevailing institutional logics and prior market reactions to the adoption of similar policies” (2004: 450-451). Therefore, prevailing institutional settings condition not only governance practices but also their financial consequences. In the United Kingdom, with its “soft” regulation and prevalence of business interconnections that result from directors’ interlocks, IPO investors view the well-networked directors as a key organizational resource that makes an important contribution to the firm’s legitimation process. In the United States, however, the prevalence of the contract-centered, agency-driven approach to corporate governance, combined with regulatory restrictions on board interlocks, may substantially reduce the salience of this governance factor within the process of stock market legitimation. In other words, the subtle differences in confluences of cognitive and normative institutions in the United States and United Kingdom identified above may have a significant effect on the perceptions of stock market “audiences” (Deephouse & Suchman, 2008).
The rationale presented in Hypothesis 2 and further developed here suggests that the interlocks of the CEO, internal board members, and external board members in the United Kingdom will reinforce each other and boost the valuation of the firm. However, in the United States, the effects of board interlocks on investor valuations will face limitations in terms of cognitive and regulatory legitimacy. Bearing in mind the importance of interlocks within the otherwise similar investor-centered approach to corporate governance in the United Kingdom, one can see that the strength of the response of the U.K. market to a particular constellation of board interlocks will be higher in magnitude than a response to a similar constellation in the United States. Therefore, we hypothesize that:
Hypothesis 3: Compared to the United States, the interlocks of the CEO, internal board members, and external board members are more positively associated with investors’ perceptions of IPO value in the United Kingdom.
Method
Sample
We employed a multistage data collection procedure, as suggested by Nelson (2003). We first included all IPOs floated on the main and secondary tier markets in each country (United States and United Kingdom) in a given time period. Specifically, we used the London Stock Exchange new issues files in the United Kingdom and Thomson Financials Security Data Corporation new issues database to identify all firms that made IPOs in the U.S. and U.K. markets between 1996 and 2010. Following prior research, we excluded readmissions and transfers, corporate spin-offs, equity carve-outs, units, and investment trusts to mitigate the problem of comparability of their organizational forms with the typical entrepreneurial firms going public (Nelson). We also excluded cross-listed IPOs from the sample as they greatly distort our institutional argument as a result of the multiple institutional contexts involved. The variables of interest came from information provided in the IPO listing prospectuses, which contain detailed information on the external directorships of board members.
Since IPO firms’ characteristics help determine their performance, we limit the impact of risk differences between U.S. and U.K. IPOs by using a matched sample of firms from each nation. This method helps ensure that, as far as possible, the sample consists of firms that match in almost all major details except for the legal institutions that they face. In line with prior research (e.g., Chahine, Arthurs, Filatotchev, & Hoskisson, 2012; Conyon, Core, & Guay, 2011; Donelson, McInnis, Mergenthaler, & Yu, 2012), a matched-pairs design allows us to better control for differences in IPO populations between the two countries. Prior studies indicate that the U.S. IPOs are, on average, larger than U.K. IPOs and more skewed towards high-tech firms than U.K. IPOs, which are dominated by low-tech firms, such as those operating in the mining, general industries, and retailing sectors (Cook, Kieschnick, & Van Ness, 2006; Reber, Berry, & Toms, 2005). Hence, using the total population of IPOs in both countries would introduce a significant sample bias.
To generate the matched-pairs methodology, we first matched firms from the U.S. and U.K. IPO samples on the basis of the Standard Industrial Classification three-digit codes. We then matched firms on the basis of the closest date of listing within the 1-year range of difference and selected IPOs with the closest possible size (market capitalization) within the 25% range of difference, which are criteria usually used as control variables in the IPO literature (Chahine & Filatotchev, 2008). Finally, we excluded IPOs where the timing of the CEO and board director appointments was unclear. The remaining sample includes IPOs with CEOs who founded or joined the firm prior to other board members. The result is a final sample of 410 IPOs (a matched sample of 205 from each country).
Measures
Investors’ perceptions of IPO firm value
A significant body of research in finance and management employs proxies for public market investors’ perceptions of firm value. For example, scholars employ underpricing (i.e., the ratio of the difference between the closing stock price on the first day of trading and the offer price over the offer price) as a proxy for market performance of the IPO (Certo et al., 2001). Underpricing represents a direct wealth transfer from the founders and initial shareholders to new investors. Since underpricing is usually skewed, empirical tests hereafter use the natural logarithm of (1+Underpricing). Nelson (2003) argues that scholars can measure investor optimism about the prospects of an IPO firm by the percent price premium, which represents the difference between the offer price and the book value at the time of the IPO and is equal to the offer price per share minus the firm’s book value per share all divided by the offer price per share at the time of IPO. A higher IPO premium would thus suggest a higher investors’ perception of the IPO’s value. Therefore, we used both IPO underpricing and IPO price premium as proxies for investors’ perceptions of IPO value.
CEO, internal directors, and external directors’ interlocks
We identified the names of board members from the listing prospectuses. We grouped these officers into three broad categories: the CEO, internal directors, and external directors. Following prior research, we calculated board members’ interlocks as their involvement in boards outside the IPO firm (Higgins & Gulati, 2003, 2006). Following Cohen and Dean (2005), we used the “cumulative interlocks” (e.g., a sum of external board memberships for a particular type of board member) as a proxy for the extent of interlocks of a particular group of directors. We calculated directors’ interlocks by using information on the board positions held in firms at IPO and over the last 5 years before the IPO, which we obtained from the Other Directorships section of the prospectus. Although earlier information is sometimes available in U.S. prospectuses, a 5-year period is the standard cutoff date used in listing prospectuses in the United Kingdom. As a robustness test, we also used the interlock intensity variable measured as the total number of interlocks over the number of directors for both internal and external directors and obtained similar results, which remain available upon request.
Controls
We included firm size and age as control variables since they may affect investor valuations (Mikkelson, Meganpartch, & Shah, 1997). We measured the IPO’s size by the natural logarithm of the firm’s capitalization at the offer price (Ln Market Cap). Since there is typically more information available about larger firms, such firms should have higher valuations (Brav & Gompers, 1997). In addition, we controlled for firm age as measured by the number of years elapsed between the firm’s founding date and its IPO date (Firm Age), and we expected older firms to have higher valuation, which is in line with firm size. We also employed a high-tech (Hi-tech Dummy) variable to control for the presence of greater asymmetry of information in high-tech firms since such IPOs are likely to underperform following their public offerings. This dummy variable equals 1 if the IPO firm is a high-tech firm and 0 otherwise. 3 We controlled for previous operating performance in the regression models related to internal and external directors’ interlocks as the IPO firm’s ability to attract directors with higher interlocks may depend on its prior performance. We used the return on assets (RoA) over a 1- and 2-year period prior to the IPO date (RoA-1 and RoA-2) as proxies for operating performance. In line with prior IPO literature, the regressions run for IPO valuations also control for operating performance prior to the IPO date (Chahine et al., 2012). Internal and external directors’ interlocks are also likely to be affected by the IPO’s risk, and we controlled for the number of Risk Factors obtained from IPO prospectuses. We expect the number of risk factors to positively affect directors’ interlocks and to negatively affect IPO valuation (Arthurs, Hoskisson, Busenitz, & Johnson, 2008).
In terms of board characteristics, we included a Board Size variable measured as the total number of non-CEO directors as larger boards are likely to have more interlocks. A founder’s leadership (e.g., a founding CEO) may affect the IPO firm’s performance (Certo et al., 2001; Nelson, 2003). Therefore, we employed a CEO Founder Dummy, which is equal to 1 if the founder is the CEO and 0 otherwise. Prior research approximates both CEO and insider ownership with alignment of their interests among public market investors, which will likely reduce underpricing and boost valuations (Certo et al.; Filatotchev & Bishop, 2002). We added both Post-IPO CEO Ownership and Post-IPO Insider Ownership, measured as the percentage of the total number of ordinary shares retained by CEOs and insiders after the IPO, respectively. Venture capitalists (VCs) possess substantial decision-making rights that may affect the governance of their portfolio companies (Barry, Muscarella, Peavy, & Vetsuypens, 1990). Therefore, we controlled for Post-IPO VC Ownership, measured as the percentage of the total number of ordinary shares retained by VCs after the IPO. We also included the CEO Age and CEO Tenure variables in the regressions run for interlocks as they may affect the extent of CEOs’ interlocks. Furthermore, we added a CEO Family Dummy equal to 1 if the IPO firm includes another shareholder, board member, or top management team member who is a CEO’s family member and 0 otherwise to reflect the extent of CEO control over the board. Given that prior IPO research does not provide clear evidence on the effects of CEO age, tenure, and presence of CEO family members on the board on IPO performance, we excluded them from the valuation regressions.
Research also suggests that the quality of the underwriter may positively affect the IPO firm’s valuation because investment banks may certify the quality of young firms during the floatation (Beatty & Ritter, 1986). Underwriter Reputation is a continuous variable that we calculated on the basis of the Loughran and Ritter (2004) rankings on a scale from 0 to 9, where 0 represents the lowest and 9 represents the highest rating. Specifically, we used the Loughran and Ritter ranking for the United States and then constructed a similar ranking for the United Kingdom for the entire IPO population.
Stock market conditions vary with time, and during some periods, IPO investors exhibit periodic overoptimism. To control for time effects on IPOs, we employed a Market Return variable, which we calculated as a weighted average of the buy-and-hold returns of the Alternative Investment Market (AIM) Index or the Main Index in the United Kingdom 4 and the Equally Weighted Center for Research in Security Prices (CRSP) Index in the United States during 3 months before the IPO date. The weights were equal to 3 for the 1st month, 2 for the 2nd month, and 1 for the 3rd month before the offering; we divided the weighted sum by 6 (see Chahine & Filatotchev, 2008). In line with Gompers and Lerner (2000), who show that periods of high funds inflow can affect firm valuations, we controlled for the technology bubble that occurred during the studied period by using a Bubble Dummy variable, which is equal to 1 if the IPO firm went public in 1999 to 2000 and 0 otherwise. Stock market momentum and underwriter reputation tend to affect investors’ perceptions of firm value at the time of IPO; however, these factors are less likely to affect the choice of board members. Therefore, we employed underwriter reputation, market return, and bubble period dummies in the IPO performance regression only, and we used other control variables in all other tests. Finally, Moore et al. (2012) indicate that IPO firms’ characteristics and their performance outcomes may differ between the U.S. and U.K. markets. To control for these differences, we also included a U.S. Dummy in the regression analyses.
Endogeneity and Instrumental Variable Analysis
Our theoretical model suggests that the extent of CEO interlocks may affect the extent of interlocks of both internal and external directors. The examination of interlocks, then, may be associated with potential endogeneity problems related to the interdependence between the interlocks of internal and external directors and their impact on valuations. To overcome these problems, we allow for joint determination of both types of directors’ interlocks by using the instrumental variable (IV) method with instruments for endogenous variables (see Chahine et al., 2011, for discussion). Specifically, we accounted for potential endogeneity by using IVs for internal and external directors’ interlocks in Equations 1 and 2 below and then used the fitted values of each of these variables in the regressions run for IPO valuation (Equation 3). To construct our instruments, we calculated the total number of media citations for all internal directors and external directors by using the Factiva database during the 5 years prior to IPO date (Internal Directors Citations and External Directors Citations, respectively). We assumed that directors of IPO firms with numerous interlocks may attract more media visibility and vice versa. However, we do not have theoretical reasons to believe that past citations related to directors directly affect IPO performance. Therefore, we construct the following system of simultaneous equations:
Given the count nature of the interlock data, we ran Equations 1 and 2 as negative binomial regressions (Models 1a and 2a reported in Table 2 discussed in the Results section below) to test the simultaneous determination of external and internal directors’ interlocks. These negative binomial regressions generated fitted values of directors’ interlocks that we used in second-stage negative binomial regressions, which had the same structure as Equations 1 and 2. Using the IV method, we employed these regressions to test Hypothesis 1 and Hypothesis 2. In addition, we used the fitted values in the ordinary least squares (OLS) regression of Equation 3 to test Hypothesis 3:
This approach helps remove the simultaneity problem and generates consistent estimators of board characteristics and performance (Brooks, 2002).
Empirical Results
Table 1 provides descriptive statistics in terms of means, standard deviations, and correlation coefficients between studied variables. These results show an average underpricing of 26.2% and an average IPO premium of 0.679 for the entire sample. In terms of board interlocks, CEOs sit on three external boards on average, whereas internal and external directors sit on 6.1 and 17.5 outside boards on aggregate, respectively.
Correlation Matrix
Note: The table uses Pearson’s (point biserial) correlation coefficients for continuous (dichotomous) variables. BoD = board of directors; IPO = initial public offering; RoA = return on assets; VC = venture capitalist.
p < .05.
p < .01.
p < .001.
An average IPO firm has a market capitalization of $411.34 million and goes public 11.9 years following its inception. Almost half of the IPOs are high-tech firms (44.1%), and 48% of them are led by their founders–CEOs. Table 1 also exhibits negative correlation coefficients between underpricing and the interlocks of CEOs, internal directors, and external directors, in line with our expectations. Moreover, the evidence indicates a positive correlation between IPO premium and the interlocks of CEOs, internal directors, and external directors. 5 Furthermore, Table 1 shows that our instruments are positively correlated with their related endogenous variables (e.g., director interlocks) but not with both proxies for IPO valuation.
Table 2 includes the negative binomial and their second-stage regression runs for external and internal directors’ interlocks. Specifically, Models 1a and 2a present the negative binominal regression results for both variables employed to generate fitted values, whereas Models 1b and 2b present the second-stage regressions, using the fitted values and controlling for the simultaneous determination of both variables.
Building Interlocks of the Board of Directors: A Simultaneous System
Note: The table controls for the potential endogeneity of internal and external directors’ networks using the instrumental variable method, in which we estimate negative binomial regressions of both internal and external directors’ network variables as first-stage regressions (Models 1a and 2a). We then use the fitted or predicted values of each of these variables in Models 1b and 2b. White heteroskedasticity-consistent standard errors and covariance. Standard errors are shown in italics. N = 410. RoA = return on assets; IPO = initial public offering; VC = venture capitalist.
p < .1.
p < .05.
p < .01.
p < .001.
The negative binomial regressions in Models 1a and 2a indicate a positive association between external and internal directors’ interlocks (at the 0.1% level). Using the Hausman (1978) test of exogeneity to verify whether both interlocks are jointly determined, we confirm that both of them are endogenously related (at the 1% level). Therefore, the residual “unexplained” terms in both Equations 1 and 2, є1 and є2, are correlated; the correlation coefficient is negative and equal to –.37 (at the 1% level). This finding suggests that the negative binomial estimates of both internal and external directors’ interlocks in Models 1a and 2a are biased, and their simultaneous relationship would later affect their IPO performance. 6 Further investigations in Models 1b and 2b include a simultaneous estimation of the two structural equations of external and internal directors’ interlocks. On the basis of the IV method, Models 1b and 2b refer to a second-stage approach to control for the endogenous determination of both variables and use their fitted values calculated on the basis of first-stage Models 1a and 2a. Regardless of the model, and in line with Hypothesis 1, both internal and external directors’ interlocks relate positively to the CEOs’ interlock (at the 0.1% level).
Table 3 provides the results of the effects of country differences on the association between external and internal directors’ interlocks. It verifies that both interlocks exhibit different associations according to the IPO country and, thus, supports our institutional framework. To test for country differences, we split CEO interlocks into two country-specific variables: U.K. CEO interlocks and U.S. CEO interlocks for the U.K. and U.S. IPOs, respectively. Table 3 indicates that both external and internal directors’ interlocks relate negatively to CEO interlocks in the United States (at the 1% level or more), whereas the interlocks relate positively to CEO interlocks in the United Kingdom (at the 0.1% level). Thus, U.S. CEOs with high interlocks are less likely to be associated with external and internal directors with high levels of interlocks, in contrast to U.K. CEOs. As a result, the findings support Hypothesis 2.
Country Differences in Building Interlocks of Board Directors
Note: The table controls for the potential endogeneity of internal and external directors’ networks using the results of the instrumental variable regressions in Table 2. The first-stage regressions in Models 1a and 2a in Table 2 provide fitted values used in the second-stage negative binomial regressions in Models 3 and 4. White heteroskedasticity-consistent standard errors and covariance. Standard errors are in italics. N = 410. RoA = return on assets; IPO = initial public offering; VC = venture capitalist.
p < .1.
p < .05.
p < .01.
p < .001.
Table 4 presents the second-stage OLS regressions for underpricing and IPO premium, using the fitted values of internal and external directors’ interlocks calculated in the first stage of Models 1a and 2a in Table 2. Models 5 and 7 examine the association between IPO valuations (underpricing and IPO premium) and various interlocks, whereas Models 6 and 8 control for the country differences. Again, to test for country differences in the effects of internal and external directors’ interlocks, we used the same variable split procedure as we used with the CEO interlocks in Table 2.
Initial Public Offering Valuations
Note: The table controls for the potential endogeneity of internal and external directors’ networks using the fitted values generated in Table 2. White heteroskedasticity-consistent standard errors and covariance. Standard errors are shown in italics. Within each row, values with different subscripts are significantly different at the 1% (a) and 5% (b) level, respectively. N = 410. RoA = return on assets; IPO = initial public offering; VC = venture capitalist.
p < .1.
p < .05.
p < .01.
p < .001.
Model 5 indicates that underpricing is negatively related to CEO interlocks, and to external directors’ interlocks, but indicates no significant effect of internal directors’ interlocks. Model 6 controls for the differential country effect and shows a negative association between underpricing and the CEO interlocks in both the United States and the United Kingdom (at the 5% and 0.1% level, respectively). The negative association between underpricing in U.S. versus U.K. CEO interlocks is significantly different at the 1% level. Moreover, underpricing is negatively related to internal directors’ interlocks in the United Kingdom at the 0.1% level and to U.K. external directors’ interlocks at the 1% level, and there is no significant association between underpricing and the interlocks of internal and external directors in the United States. This finding suggests that building interlocks in the United Kingdom adds value to IPO investors, which is in line with Hypothesis 3.
Empirical results in Models 7 and 8 using IPO premium as a dependent variable confirm the results in the regressions in Models 5 and 6. Specifically, Model 7 shows that IPO premium is positively related to CEO interlocks (at the 0.1% level), and the results in Model 8 are less significant in U.S. compared to U.K. IPOs (10% and 0.1%, respectively). This finding suggests that firms benefit from the interlocks and connections provided by their CEOs in both the U.K. and the U.S. IPOs, and the positive association between IPO premiums in U.K. versus U.S. CEO interlocks is significantly higher at the 5% level. Model 7 shows that the IPO premium is positively related to internal directors’ interlocks (at the 0.1% level), and the results are significant only in U.K. IPOs in Model 8. This finding suggests that outside investors are likely to pay a higher premium for U.K. IPOs with a greater interlock of internal directors than is the case in the United States. Similarly, the interlocks of external directors increase IPO premiums in Model 7, and the results in Model 8 are significant only in U.K. IPOs (at the 1% level). Again, the external directors’ interlocks appear to significantly affect the IPO premium paid in U.K. IPOs. Therefore, the evidence supports Hypothesis 3 and shows that, compared to interlocks in the United States, the interlocks of the CEOs, internal board members, and external board members in the United Kingdom are more likely to improve investors’ perceptions of IPOs. 7
We also performed a number of robustness tests. Our empirical investigations used the total internal and external directors’ interlocks, but this approach may create a problem with skewed distribution of interlocks within the boardroom, whereby some directors may have extensive external interconnections while others do not. Therefore, in the leftmost columns in Table 5, we calculate the proportions of directors with external interlocks and use this instead of the “cumulative interlocks” variables. The results are in line with the main findings: The greater the proportion of directors with interlocks, the lower the underpricing and the higher the premium, and this effect is more significant in U.K. IPOs.
Different Types of Directors’ Interlocks and Initial Public Offering Valuations
Note: White heteroskedasticity-consistent standard errors and covariance. Standard errors are in italics.
p < .1.
p < .05.
p < .01.
p < .001.
Furthermore, we base our empirical evidence on board interlocks regardless of whether they are within the same industry. Therefore, as an additional robustness test, we repeat the analysis in Tables 3 and 4 using CEO’s, internal directors’, and external directors’ interlocks within similar industries to the IPO firm with results reported in the middle columns of Table 5. The results are also consistent with our predictions and confirm the significant effect of directors’ interlocks on IPO performance in the United Kingdom. The perceptions of the importance of external board connections among investors may be particularly salient for board interlocks with firms in financial services. Therefore, IPO firms may want to appoint directors with strong connections in finance and accounting. Therefore, we repeat our analysis focusing on the proportion of directors with interlocks in the financial industry, reported in the rightmost columns of Table 5. These data show evidence of the positive effect of this type of directors’ interlocks on IPO performance, mainly IPO premium, in the United Kingdom. Compared to our main results in Tables 3 and 4, the data on directors with similar industry interlocks and directors with financial industry interlocks shown in Table 5 suggest that industry and finance experiences of IPO board members associated with their interlocks are important but are not the only channels of the overall legitimation mechanism. 8
We measure our interlock variables as the sum of external board memberships of different board participants, and they say very little about how centrally located the IPO firm is in the overall board network structure and possible overlaps of the CEO interconnections with that of boards (e.g., common ties, strong ties, indirect ties). Although a comprehensive network analysis of board linkages in the two populations of firms is not possible as a result of data constraints, we performed further robustness tests by using available data on the board members’ characteristics and external connections. 9 First, we eliminated from the insider and outsider ties (interlocks) all ties that overlap with CEO ties. This approach means that the CEO ties variable remained the same, whereas the ties of insider and outsider board members decreased by excluding the overlaps with the CEO. The regression results using new variables confirm our key findings.
Second, building on research by Renneboog and Zhao (2011), we calculated a measure of the IPO firm centrality (the betweenness score). We employed this measure as a regressor in the multivariate regressions at the firm level with IPO performance as the dependent variable. The results showed that U.K. firms are more likely to benefit from their “intermediary” position within the overall IPO board network to generate higher valuations compared to U.S. firms. This finding is in line with our discussion of the possible impact of the City of London network economy on IPO valuations in the United Kingdom compared to the United States.
Third, our focus so far was on the extent of interlocks of the CEO and board members and their valuation outcomes. Another way of looking at IPO board connections is to consider them as CEO ties but with different degrees of strength to better understand the role of the CEO’s position within the IPO firm’s network. Following Granovetter (1973), we considered CEO direct interlocks that include their personal involvement in other boards as strong ties. The CEOs’ indirect interconnections with other actors through other board members in the IPO we considered as indirect ties. However, we expect CEOs’ indirect interlocks, provided through IPO firms’ board members who have close relationship with CEOs, to be “stronger” indirect ties than the remaining “weak indirect ties” provided by other board members. This close relationship is based on previous experience in the same firm, similar nationality, and/or similar professional memberships (certified public accountant, chartered financial analyst, etc.). We excluded education because available data on the education of board members in the U.K. firms are limited. As such, we divided weak ties into “strong indirect ties” through closely related directors and “weak indirect ties” provided by the remaining directors. We used these new variables as regressors in our analysis of IPO valuations. The results show that CEO strong ties are more likely to increase performance in U.K. IPOs. Moreover, CEO strong indirect ties are more likely to add value than CEO weak indirect ties. Again, the effects on performance of U.K. CEO strong indirect ties were more significant than those of U.S. CEOs. Interestingly, the U.K. CEOs’ weak indirect ties still add value (lower underpricing and higher premium), which is not the case in the United States.
Finally, given the important role VCs play in guiding the selection of new employees and board members before an IPO, we repeated our tests by using a sample of 156 non-VC-backed firms only; the results remain significant, particularly in the case of U.K. board members. Therefore, we are confident that the VCs alone do not drive board member selection.
Discussion
Scholars have built much of corporate governance research in the context of IPOs on the agency-grounded assumption that governance factors may have an impact on information asymmetries in an IPO listing and, therefore, inform external investors in terms of their valuation decisions. For example, scholars typically see variables related to board characteristics, such as CEO duality, board independence, and directors’ prestige, as separate causes of IPO success, and these governance practices have an ability to influence the firm’s stock market performance. Yet the academic evidence on the actual valuation implications of board-related governance issues is inconsistent.
We challenge the basic assumptions of past research grounded within the agency perspective both by focusing on the workings of legitimacy associated with the IPO’s board member interlocks and by proposing important theoretical extensions building on research within the field of institutional theory. We provide insight into the legitimization mechanisms that guide the IPO firm’s actions in response to pressures coming from stock markets. Specifically, the CEO interlocks matter, but the importance of the CEO interlocks lead in turn to the ability to attract board members whose interconnections with other actors matter too. Our evidence indicates a positive relationship between the extent of interlocks of the board members (internal and external) and the extent of external board memberships of the CEO. This finding is in line with recent research that suggests the existence of a close relationship between adoption of logic-defining and logic-extending organizational practices within a general context of legitimation (e.g., Pollock et al., 2008; Shipilov et al., 2010).
Additionally, we extend this finding further by showing that institutional settings affect the legitimation mechanisms. Specifically, we show that the extent of directors’ interlocks in the United Kingdom is more strongly related to CEO interlocks than in the United States. Such a finding, we argue, is consistent with the institutional perspective on corporate governance that considers governance factors as drivers of the firm’s legitimacy among the stock market participants (Bell et al., 2014; Moore et al., 2012). Our empirical results show that CEOs’, internal board members’, and external board members’ interconnections with other actors are more important to IPO valuations in the United Kingdom than in the United States. Therefore, the results highlight that in the United Kingdom, with its network-based model of corporate governance, interlocks of the CEO and of the external and internal board jointly play a role in building confidence in the firm and its corporate governance. However, in the United States, with its reliance on the structural board characteristics to guide corporate governance, interlocks of board members may be less critical, a finding we confirm by numerous robustness tests. Therefore, we extend previous studies on organizational legitimacy by showing that the process of legitimation is far from being universal across national capital markets, and it may occur only in a specific institutional environment.
The findings here also expand our understanding of the institutional perspective on corporate governance by addressing institutional multiplicity. The United States and the United Kingdom share a strong bond in their common law foundation and focus on shareholder value as a guiding principle of firm-level governance. However, the two nations have different confluences of regulatory, normative, and sociocognitive institutions that may have a significant bearing on corporate governance. The United States relies on a strong legal foundation with strict laws and rules. In contrast, the United Kingdom relies on a network-based model of corporate governance in which trust and strong social relations among the parties play a significant role. As a result, external connections of the board members in the United States and the United Kingdom play different roles in relation to the stock market participants’ expectations. Thus, the institutions of a nation play a critical role in shaping the firm-level corporate governance and organizational outcomes of various governance practices. As scholars moving forward, we need to be aware of the impact of institutions on corporate governance. Too often, scholars have looked to single nations and used the information gained to argue that their insights have worldwide implications. The evidence here is that scholars need to pursue more fine-grained analyses in which they consider individual institutions.
Finally, the research here is highly relevant in terms of practical debates with regard to optimal board characteristics, especially in relation to companies seeking prestigious board members. Some scholars argue that multiple demands on board members who serve on too many boards can distract them and reduce the value they provide to the firm. 10 In addition, appointing well-connected board members creates internal, nonfinancial costs since CEOs and other insiders may perceive adding a prestigious new director as a threat to their power or standing in the firm’s status order (Acharya & Pollock, 2013). However, our evidence indicates that, in the high uncertainty environment of an IPO, investors value the multiple board memberships that directors bring to the venture. Thus, rather than stretching the board members too thinly, such services create value for the IPO firm. More importantly, this impact of external memberships is context specific, and scholars as well as practitioners need to conduct a thorough analysis of institutional confluences the firm is embedded in to find an optimal configuration of the board.
Limitations and Future Research
Several limitations in our analysis create opportunities for scholars to expand on the foundation built here in future research. The first is that we focus on a single dimension of corporate governance—board interlocks. A rich range of other network variables offers potential sources of future examination of the legitimation process of an IPO. For example, in the future, scholars should examine the belief that board members’ political connections and informal, social ties will result in better performance (Chahine & Goergen, 2013; Chen et al., 2008; Filatotchev & Bishop, 2002). While the study of IPO outcomes outside of the United States is a good first step, we need many more studies in multiple governance environments involving developed and developing economies. Clearly, the role of interconnections is particularly important in emerging economies (Peng & Luo, 2000). The expansion across multiple nations should include the examination of both developed and emerging economies.
Our use of cross-sectional data also limits the research. The IV methodology helps us to deal with endogeneity; however, we are still unable to explore the impact of time on our results. Future research should expand on our investigation longitudinally to better understand the impact of time on a firm’s responses to institutional pressures to conform. Future research should also seek to use qualitative methodologies to explore a richer set of interconnections that may exist among the various parties, including key investors and others that could affect the legitimacy of the firm. Interconnections among actors, such as the time spent on board deliberations, where that time is spent, and how the various board members and investors connect both financially and socially, may be as important to ultimate IPO valuation.
Conclusion
Scholars increasingly recognize that institutions play a critical role in putting isomorphic pressures on firms. This research helps to contextualize the understanding of the legitimation process by examining the role played by external linkages of directors in IPO firms to other legitimate entities. Overall, our study provides a more complex picture of the governance-performance relationship than traditional agency-grounded research. We demonstrate that IPO firms may achieve legitimacy by attracting directors with substantial external interlocks. However, this process of legitimation and its value outcomes nest within a broader institutional framework that takes into account the firm’s institutional environment.
Footnotes
Acknowledgements
This article was accepted under the editorship of Patrick M. Wright.