The Tradeoff Between Private Equity Sponsorship, Board Centrality, and Experience as Credible Signals for IPO Performance

Abstract

Private equity funds implement various management and governance practices in firms they endorse, signaling higher quality of sponsored IPOs (Initial Public Offer). However, the participation of such PE funds comes at a cost for newcomer firms, as they may lose both autonomy and future post-IPO earnings. If they do not choose to signal quality through PE funds, the IPO literature points to the validity of other mechanisms, such as the experience and centrality of the board of directors. We theorize and test the effect of the tradeoff between private equity sponsorship and board centrality and experience on IPO performance. We analyzed the effects of signaling on long-term performance through the one-year cumulative abnormal return in Brazilian IPOs issued between 2004 and 2013. Our results indicate that private equity sponsorship is the most effective signal, completely overtaking the relevance of board centrality and experience to explain IPO performance. However, these latter board characteristics significantly affected the performance of non-private equity-backed IPOs. Between PE-backed companies, there is a reverse effect of the board centrality on the IPO's performance and a substitution effect of the board's previous experience with IPO processes.

摘要：

私募基金通常对其投资的公司采用各种管理和治理措施，来表现这些公司上市时的高质量。但是，私募基金的参与同样会对这些公司造成损失，一是自主权的丧失，二是上市之后的盈利的分配比。假如不选择通过私募基金来给出高质量的信号，IPO文献指出可以用其他机制来达到同样的目的：即董事会成员的经验和在业界的中心化程度。本文意在探索在预测IPO绩效时这三者之间的得失。我们对2004-2013年巴西所有的IPO进行了分析，通过其一年的累计异常回报来检验信号效应对公司长远绩效的影响。结果表明，私募基金赞助的信号是最有效的，完全取代了董事会经验和中心化程度的影响。但是，后两个因素显著影响了非盈利私募基金投资的IPO公司。此外，在私募基金投资的公司中，董事会的中心化程度对IPO业绩起了相反的作用，董事会成员以往在IPO过程中的经验对IPO业绩起到了替补的作用。

INTRODUCTION

An Initial Public Offer (IPO) is among the most critical events a firm goes through. It constitutes an opportunity for newly opened firms to gather resources in the market (Li, Vertinsky, & Li, 2014; Pagano, Panetta, & Zingales, 1998). Yet, it also poses challenges and burdens (Li & Yao, 2019). Firms must attain legitimacy among potential investors, while the latter is unable to directly observe the extent that the firm has properly prepared for the IPO (Courtney, Dutta, & Li, 2017; Hsu & Ziedonis, 2013; Reuer, Tong, & Wu, 2012). This asymmetry of information requires that firms send credible signals to the market (Certo, 2003; Connelly, Certo, Ireland, & Reutzel, 2011; Spence, 1974). Recent literature has explored how different signals are interpreted by investors, as well as the contingencies involved in their effectiveness. Thus, how and why certain signals are chosen and not others have been the focus of recent scholarship (Kleinert, Bafera, Urbig, & Volkmann, 2021).

While we embrace this debate, we posit that this research agenda has devoted insufficient attention to firms embedded in emerging economies (Chakrabarti, 2019). Within emerging economies, firms’ association with institutions leads to higher odds of survival and superior performance (Li & Yao, 2019). These institutional ties comprise relations to State and public agencies, as well as affiliation under economic groups (e.g., Hoskisson, Eden, Lau, & Wright, 2000). Within emerging economies, these institutional ties provide access to scarce resources and offer buffering against economic upheavals (Musacchio & Lazzarini, 2016). Further relevant to the IPO process, local institutions’ specificities present important challenges for managers who lack specific expertise (Lazzarini, 2010). For that reason, IPO firms might signal higher odds of success by attracting experienced directors to its board (Cho & Arthurs, 2018). Thus, by attracting central and experienced directors, IPO firms are able to signal higher odds of survival to potential investors. At the same time, firms embedded in emerging economies might try to be associated with private equity firms, in order to signal high competitiveness to potential investors (Minardi, Ferrari, & AraújoTavares, 2013). We posit that these distinct choices are not simply additive, as not all IPO firms will be able to attract PE funding. Along the same line, attracting central and experienced board members might constitute a second-best alternative. As a result, IPO firms within emerging economies face a tradeoff between associating with central and experienced board members or attracting investments from private equity firms. Thus, this tradeoff suggests a gap in the literature that can be stated as: how do IPO firms embedded in emerging economies decide between distinct signals to potential investors?

In order to answer this research question, we chose to explore the effectiveness of diverse signals from IPO firms to investors in the Brazilian market. In the early 2000s, Brazil was among the most prominent emerging economies. Before that period, Brazil underwent several liberalization processes, including a wave of privatization of State-owned companies. During the timeframe analyzed, Brazil was in the spotlight for its stream of IPOs, which attracted the attention of many national and international private equity firms. At the same time, Brazil's capitalism still presented several features that were frequently associated with obstacles to development (Lazzarini, 2010). Brazil is characterized as associated to ‘state as minority investors’ type of capitalism, where the state holds only minority equity interest in several major public firms and is perceived as interventionist (Musacchio, Lazzarini, & Aguilera, 2015; Wright, Wood, Musacchio, Okhmatovskiy, Grosman, & Doh, 2021). This means that although the State had reduced its direct and indirect participation in the economy, it was still expected that it would intervene to rescue failing firms perceived as ‘national champions’ (Wright et al., 2021). Consequently, ties to government officials and politicians could generate benefits to firms engaged in their IPO process. As a result, Brazil's IPO market during the 2000s constituted an attractive context to explore the tradeoff firms experience while choosing appropriate signals to potential investors.

This study contributes to the private equity literature in emerging markets (Cho & Arthurs, 2018), especially regarding the effect of different IPO quality signaling mechanisms (Bell, Filatotchev, & Aguilera, 2014). We demonstrate that a company's association with a private equity firm works as a credible sign for IPO quality, since it likely increases the odds of positive IPO performance. Second, we show that in emerging markets with weak legal protection, private equity funds are the most relevant signal (Bruton, Filatotchev, Chahine, & Wright, 2010; Lee, Pollock, & Jin, 2011), obliterating the impact of other variables related to the directors’ experience and board centrality. Third, when companies lack PE investment, the search for more central and experienced directors conducting IPO processes are credible positive signals to the market (Cho & Arthurs, 2018; Field & Mkrtchyan, 2017; Filatotchev, Chahine, & Bruton, 2018; Reuer et al., 2012). Fourth, following other studies that used specific measurements of experience (Cho & Arthurs, 2018; Field & Mkrtchyan, 2017; McDonald, Westphal, & Graebner, 2008; Peruffo, Marchegiani, & Vicentini, 2018), we demonstrate how recurrent experiences in the same domain – directors’ previous experience in IPO processes, can promote greater performance. Fifth, we demonstrate that different quality signals may not be additive, pointing to substitutability (Khoury, Junkunc, & Deeds, 2013) and reversibility when analyzed together.

In this article, based on 120 IPOs in Brazil between 2004 and 2013, we propose three credible signals to launch a successful IPO. First, IPO PE (Private Equity) sponsorship can be understood as a signal of possible further gains, because private equity funds conduct a meticulous selection process before making the investment decision, improve information flows to the board; increase board control over managers; and sharpen financial incentives to directors to enhance monitoring of derivative exposure (Masulis & Thomas, 2009), thus better preparing the company for going public. Second, we test the effect of board centrality – the number of connections with other board companies – on IPO performance. Following Filatotchev et al. (2018), we support the idea that the board centrality of IPO firms signals competence, social acceptance, and prestige to investors. But at the same time, board centrality might signal the association of IPO firms to traditional economic groups and networks of political influence. Third, we note that the presence of experienced directors on the board signals to outside investors that the issuing company has an efficient corporate governance framework (Gray & Nowland, 2013; Kroll, Walters, & Wright, 2008; Payne, Benson, & Finegold, 2009). Within the Brazilian context, experienced board members might be instrumental in order to cope with specific contextual hurdles (Lazzarini, 2010). Finally, we posit and test the hypothesis that firms face a tradeoff between attracting private equity firms and choosing other available signals. We observe that, in the absence of more prominent signaling (Lungeanu, Stern, & Zajac, 2016), which is the existence of backed PE fund, companies will try to demonstrate a greater likelihood of being successful in the IPO process through other signals, such as the centrality and experience of the directors which were allocated on the board. We use the one-year cumulative abnormal return as a proxy for the credibility of the signal.

This article is organized as follows: First, we present the arguments that support our hypothesis about signaling. Second, we describe the data, define the variables, and specify the method. Third, we present the results, where we examine some additional analyses. Fourth, we discuss those results, pointing to contributions, and list the limitations and suggestions for future studies. Finally, we synthesize the conclusions of the study.

IPOs AND SIGNALING: BRIEF REVIEW AND SPECIFICITIES IN EMERGING COUNTRIES

Going public is one of the most critical decisions an organization can make. By going public, companies have access to equity and debt capital to finance their growth (Li et al., 2014; Pagano et al., 1998). The IPO enhances company image and publicity, provides advantages such as an informative stock price, liquidity for shareholders, and increases competition among finance providers (Röell, 1995). IPO also allows partial or total exit of existing shareholders and portfolio diversification of initial owners (Pagano et al., 1998; Röell, 1995).

While IPOs might enhance firms’ access to market resources, they frequently lack sufficient legitimacy among potential investors (Li & Yao, 2019). Investors might be wary about the firm's IPO performance (Cochrane, 2005; Li & Mahoney, 2011). This suspicion is motivated by the information asymmetry between the company seeking to become public, and the market that has an interest in its stocks (Courtney et al., 2017; Hsu & Ziedonis, 2013; Reuer et al., 2012).

The trajectory toward opening a firm's capital involves dealing with several roadblocks. Before going public, the company must have internal controls, systems, and procedures to report reliable financial information to investors, and a management team that is willing and able to take responsibilities of public companies over (Chahine, Filatotchev, & Zahra, 2011). That team will be involved in meetings with board of directors, auditors, shareholders, analysts, and press. They will face pressure for returns, and they will have to present feasible planning (Mayr, 2011). The company must commit to corporate governance practices, with the board of directors playing a significant role (Filatotchev et al., 2018). Good practices require independent board members and investors attribute value to board member expertise and experience, and very often require changes to the board of directors and the establishment of proper committees (Khoury et al., 2013; Mayr, 2011). From the potential investors’ perspective, these changes are necessary not only because investors require that assets yield competitive returns vis-à-vis equivalent risks, but most importantly, because investors need to be reassured that the firm has a sound and reliable business model.

The inherent conflict throughout this process is that while potential investors are eager to have the opportunity to invest in the new stock, they fear that the asymmetry of information between them and the firm's management will impose unforeseeable risks to the asset (Michaely & Shaw, 1994). For that reason, potential market investors in companies undergoing the IPO process require clear signals that its management is going to pursue all internal changes (Certo, 2003; Connelly et al., 2011).

Extant literature on ‘signaling theory’ and strategy has explored how entrepreneurs and firms signal quality to the investment market (Connelly et al., 2011; Spence, 1974). The idea of ‘signaling’ was originally developed by economists in order to take into account how actors are able to cope with the asymmetries of information in imperfect markets and signal unobservable quality (Connelly et al., 2011; Spence, 1974). Extant literature has established that signaling is credible when it is costly to firms. As a consequence, producing reliable signals entail costs that might be prohibitive for many firms (Spence, 1978).

Through signaling, the company's management team attempts to persuade its audiences that it has the required skills to run it well as a public company. These signals might be pursued throughout several mechanisms, including hiring prestigious executives (Acharya & Pollock, 2013; Chen, Hambrick, & Pollock, 2008), associating with prestigious private equity firms (Gulati & Higgins, 2003; Hallen, 2008; Minardi et al., 2013; Reuer et al., 2012) and with prominent investment banks and alliance partners (Reuer et al., 2012). Recent scholarship has explored how audiences interpret a multiplicity of signals (Paruchuri, Han, & Prakash, 2021). For instance, investors’ reaction to CSR initiatives (DesJardine, Marti, & Durand, 2021), media attention (Courtney et al., 2017; Vanacker, Forbes, Knockaert, & Manigart, 2020), private equity experience (Zhang & Yu, 2017), public agencies funding (Islam, Fremeth, & Marcus, 2018; Stevenson, Kier, & Taylor, 2021), association to prestigious venture capitals and universities (Colombo, Meoli, & Vismara, 2019), rhetoric and identity congruence (Steigenberger & Wilhelm, 2018; Yang, Kher, & Newbert, 2020), and entrepreneurs’ networks (Roma, Messeni Petruzzelli, & Perrone, 2017). Current developments in the signal theory applied to IPOs and entrepreneurship have attempted to provide more nuanced explanations, where signals’ effectiveness depend on several contingencies (Kleinert et al., 2021).

Among the possible signals firms produce before their IPOs, we draw attention to their relations to the State and economic groups as central to explain IPO performance in emerging markets. Within the context of IPOs, relationships to government and public institutions are important signals, as investors are assessing firms’ credibility (Li & Yao, 2019; Stuart, Hoang, & Hybels, 1999). In emerging markets, due to less developed markets, performance is highly dependent on government support (Hoskisson et al., 2000). Conversely, without government ties, it is harder to attain higher performance levels (Li & Yao, 2019). Thus, IPO firms attempt to establish ties to the government and public institutions to diminish uncertainty to potential investors. Within emerging markets, IPO firms might also benefit from affiliation to economic groups, as the latter is expected to provide a range of important resources, including the capital, expertise, and management (Silva, Majluf, & Paredes, 2006). Yet, IPO firms’ association with economic groups has shown ambivalent results and frequent higher underpricing. Analysts point that economic groups might be tempted to channel resources out from one firm to another, harming minority stakeholders’ interests (Chakrabarti, 2019).

Board Directors and Board Interlock

Board members play an important role in enforcing effective governance on firms. They support the implementation of managerial practices and bring information to the firm (Davis, 1996). Thus, prominent board members in the business community send ‘certification’ signals of quality (Lazzarini, 2010). Yet, board members also receive and exert influence. Thus, their activity bears an important political dimension (Davis, 1996). In the Brazilian context, interaction among board members within and across firms might contribute to firms’ collusion (Lazzarini, 2010).

Boards are comprised of directors who participate in more than one company (Davis, 1996). As such, directors have the opportunity to interact with members of several companies, serving as a communication channel. This channel is known as board interlocking, in which a director from one company also has a seat on the board of another company (Mizruchi, 1996). Interlock across firms occurs when economic groups appoint the same directors to firms that belong to the same group, or when firms choose independent board members that also sit at other boards (Lazzarini, 2010). When firms choose board members, recognition and prestige of the directors are taken into account (Acharya & Pollock, 2013; Filatotchev & Bishop, 2002). Search committees attribute high prestige to directors with a position in other companies (Ferris, Javakhadze, & Rajkovic, 2017; Haunschild & Beckman, 1998).

As companies incorporate into their board directors who are already central to other companies (Certo, 2003; Rossoni & Mendes-Da-Silva, 2018), and who also have experience, knowledge, and information (Larcker, So, & Wang, 2013; Mol, 2001), the presence of such directors signal the quality of the board (Filatotchev et al., 2018; Gulati & Higgins, 2003). That happens because it is thought to maintain a differentiated capacity of judgment (Shleifer & Vishny, 1997) and demonstrates the company has upgraded its knowledge to improve its corporate strategy (Carpenter & Westphal, 2001; Haunschild & Beckman, 1998).

In the Brazilian context, the board interlock (the set of affiliations of board members to firms) is seen as more cliquish than what is observed within other emerging countries (Brookfield et al., 2012). This is interpreted as a resistance to let new actors and stakeholders play a prominent role in the board interlock community (Lazzarini, 2010). As a consequence, firms remain relatively closed to new ideas and stakeholders (Brookfield et al., 2012; Lazzarini, 2007).

Brazil: Context and IPOs

Within emerging markets, the role of economic groups is important due to the prevalence of weak institutions and underdeveloped capital markets (La Porta, Lopez-de-Silanes, Shleifer, & Vishny, 2000; Mesquita & Lazzarini, 2009). Yet cooperation among firms within economic groups in Brazil has been frequently associated with collusion and clientelism, as economic groups frequently leverage their political ties to attain privileges (Frynas, Mellahi, & Pigman, 2006; Lazzarini, 2010).

As in other emerging markets, firms in Brazil need to establish direct or indirect relations with governmental and public institutions to weather economic shocks, hence buffer from uncertainty (Evans, 2012; Musacchio & Lazzarini, 2016; Schneider, 2009a). Yet, firms’ relationships with governmental and public agents might be exploited in order to exert influence. As such, Brazilian capitalism might also be characterized as ‘crony capitalism’ (Haber, 2013). Within this context, government and public officials suffer pressure from firms’ collective entities to grant privileges as sectoral lower taxes, favorable negotiations with labor, reduction of bureaucratic expenses. Individual firm interests are advanced as firms access politicians in order to accomplish specific gains as public investment in infrastructure that will enhance the firm's operations, better credit conditions, or support for international expansion (Lazzarini, 2010; Musacchio & Lazzarini, 2016).

It was expected that the wave of privatization during the nineties in Brazil would decrease the economy's dependency on the State while decreasing the economic groups’ stronghold. Instead, the State kept its influence through minority equity positions (Musacchio & Lazzarini, 2016). Although the transition toward the ‘State as a minority investor’ variety of State Capitalism brings the expectation of lower influence of economic groups on the government, the State still holds residual interest on several firms, leading to high likelihood of state intervention (Musacchio et al., 2015; Wright et al., 2021). Furthermore, Brazil has three conditions: ties and connections with government and economic groups are especially important to survive the turbulent economic environment, CVM (Brazilian Security Exchange Commission) requires transparency and disclosure rules of public companies that enable collecting public data about IPOs, board composition, and other companies’ characteristics, and it was one of the emerging countries that most raised capital with PE firms during the analyzed period.

Economic groups were also able to survive the liberalization wave, as they expanded to several sectors and maintained their block holding (Schneider, 2009b). Along the same line, it was expected the wave of IPOs would motivate the entrance of new players and diminish the concentration around traditional economic groups. Until 2007, Brazil was in the spotlight of IPOs (Di Miceli da Silveira, 2014). Yet, many firms’ boards going through IPOs reported being quite embedded into the pre-existing board interlock (Lazzarini, 2010). Financial institutions (including public banks) were able to influence newly opened firms through ownership and board members, as these banks retained ownership after lending funds to such firms (Santos et al., 2009). At the same time, many firms going through IPOs were supported by Private Equity investments. Thus, private equity firms emerged as important players, bringing their own set of managerial practices, skills, and personnel. As mentioned before, private equity practices are frequently associated with higher transparency levels, accountability, and meritocracy (Lazzarini, 2010).

HYPOTHESES DEVELOPMENT

Private Equity Investment and Long-Term IPO Performance

Exiting the investment is a crucial part of the private equity cycle and IPO and sale through acquisition are considered the most successful alternatives (Johan & Zhang, 2016; Zarutskie, 2010). IPO contributes to building the reputation of fund managers (Megginson & Weiss, 1991). A private equity manager with a successful track record signals to the market that its invested companies represent exceptional investment opportunities (Goktan & Muslu, 2018). As private equity funds are constantly bringing companies public, the IPO performance of past-backed IPOs is an important endorsement of the fund's credibility (Chahine, Filatotchev, Bruton, & Wright, 2021). Conversely, private equity firms lose reputation when forced to delist firms previously in their portfolios (Gomulya, Jin, Lee, & Pollock, 2019). Private equity firms have limited time and resources. Hence, their funding of a firm is seen as an important signal, for external observers assume that private equity firms will only invest in potentially successful prospects (Bruton, Chahine, & Filatotchev, 2009).

Several reasons lead private equity-funded firms to achieve high post-IPO performance. Private equity funds are usually accountable for internationally managed funds. As a result, contracts between private equity firms and founders lead to the disclosing of information and diminish asymmetry (Filatotchev & Bishop, 2002; Shane & Cable, 2002). Private equity firms signal of ‘certification’ to IPOs (Daily, Certo, Dalton, & Roengpitya, 2003; Gompers & Lerner, 2004; Lee et al., 2011; Megginson & Weiss, 1991), thereby reducing the winner's curse (Rock, 1986) of uninformed investors. Firms backed by private equity are seen as better prepared to go public due to several changes in the funded firm. Due diligence conducted by the fund before making the investment decision results in selecting the best-prepared companies among numerous candidates. More sophisticated executives enhance supervision, develop transparency, and improve management controls. There is a better alignment between managers and shareholders’ interests through performance reward payment. Private equity firms also improve information flows to the board, increase board control over managers, sharpen financial incentives to directors to enhance monitoring of derivative exposure, and attract highly qualified directors (Bruton et al., 2010; Masulis & Thomas, 2009; Minardi et al., 2013).

The empirical results of the relationship between private equity funds and IPO performance reflect these arguments, despite some variations across national markets. For example, in the American capital market, Gibbs and Hao (2018), in a sample of 213 IPOs between 2005 and 2006, pointed to a higher long-term return for companies backed by private equity funds compared with non-backed private equity companies. For foreign companies that go public on the US stock exchanges, both Francis, Hasan, Lothian, and Sun (2010) and Li et al. (2016) found that companies covered by venture capital funds (there is not always a distinction between venture capital and private equity in studies), tend to have higher returns. Francis et al. (2010) also highlight that the effect of companies originating from segmented markets (mostly emerging) tend to have a greater effect of venture capital funds on performance than in integrated (non-emerging) markets.

In other consolidated markets, Bruton et al. (2010) identified that venture capital funds have a significant effect on the performance of IPO in the English market, but not in the French one. For the authors, ‘VCs in the UK are significantly more likely to implement US-style IPO monitoring based on formal contractual terms including liquidation preferences, antidilution protection, vesting provisions and redemption rights, and others, as compared to their continental counterparts’ (Bruton et al., 2010: 506). This result is corroborated by Bergströn, Nilsson, and Wahberg (2006) and Levis (2011), who find not only a significant effect on the long-term performance of the IPO in the London market but also in the Parisian one, despite the effect being smaller in the latter.

In emerging markets, the positive effects tend to be more significant and consistent. For example, in India, Drebinger, Rai, and Hinrichs (2019) point out that private equity-backed IPOs perform better than non-backed IPOs both in the short and long run. Sivaprasad and Dadhaniya (2020) find a similar result both regarding operational performance and stock prices. In China, where there is strong state influence on the capital market, a hybrid type of private equity – private placement of public equity securities, also known as private equity placement (PEP) – has shown a significant effect on the long-term performance of firms where such funds are participants (Dong, Gu, & He, 2020). Johan and Zhang (2016), in a sample of 2,733 firms backed by private equity funds in 35 different emerging countries, identified that firms covered by such funds can mitigate the potential costs associated with inefficient and corrupt business environments, reinforcing the role of reducing the selection of adverse problems investment.

Finally, in the Brazilian capital market, from the 2000s onwards, all evidence points to a positive effect of the participation of private equity funds in the performance of the IPO. Minardi, Kanitz, and Bassani (2014) highlight that private equity funds in Brazil perform better than American funds. For Rossi and Martelanc (2013), this is because the practices of such funds tend to create value for the covered companies, mitigating the effects of information asymmetry, signaling good conduct for market agents. So much so that Minardi et al. (2013) identified significant abnormal returns in favor of companies covered by private equity funds, as well as Sincerre, Sampaio, Famá, and Flores (2019), who found that such companies have higher sales growth and long-term profitability.

Consequently, given such arguments and empirical evidence, we propose that PE sponsorship can work as a credible signal on the IPO quality before issuance. We posit that:

Hypothesis 1: IPO companies associated with private equity firms will have higher IPO performance than companies not associated with PE.

Board Centrality and Long-Term IPO Performance

Extant literature has emphasized ‘three distinct measurements of legitimacy that board centrality of IPO firms signals to investors (Deephouse & Suchman, 2008; Filatotchev et al., 2018): a pragmatic one, that refers to the competence of the directors; another moral, which indicates the social acceptance of board members of one company by others; and the last, cognitive, that deals with “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’ (Filatotchev et al., 2018: 1625). Further, firms with higher centrality can access higher levels of information and resources (Brass, 1984; Krackhardt, 1990; Park, Zhang, & Keister, 2019). Accordingly, there is evidence of the impact of board centrality on IPO performance (Certo, 2003; Chahine et al., 2011; Chahine & Goergen, 2013; Echols & Tsai, 2005; Filatotchev et al., 2018).

These explanations associating board centrality and firm legitimacy must be expanded to account for emerging economies’ specificities. Several developing countries suffer from ‘institutional voids’, leading to weak guarantees to investors and underdeveloped capital markets (Gao, Zuzul, Jones, & Khanna, 2017). As a consequence, in many of those countries, board centrality (commonly operationalized as the number of interlocks) functions as conduits of resources, substituting and complementing the function usually assumed by developed stock markets (Shaw, Cordeiro, & Saravanan, 2016). Brazil follows the pattern of other emerging markets where board interlocks are instrumental in providing access to economic groups, State, and public agencies (Mesquita & Lazzarini, 2009). This access is frequently perceived as instrumental for survival, even though in Brazil the State's role has shifted toward a ‘minority investor’. In addition to the resources usually accessed through board members, firms expect that these relationships will grant access to networked resources, including scarce credit, privileged information, and bailout against economic crises (Lazzarini, 2010). Research on the Brazilian market shows that political connections increase equity valuation (Bandeira-de-Mello, Arreola, & Marcon, 2012; Claessens, Feijen, & Laeven, 2008). Analogously, extant literature shows that board centrality in the Brazilian market has a positive and significant effect on the performance of the companies (Mendes-da-Silva, Rossoni, Martin, & Martelanc, 2008; Rossoni & Mendes-Da-Silva, 2018).

Therefore, board centrality works as a signal because it indicates that the company has a governance practice level that is good enough to be able to hire and accept reputable board members and that the linkages and access to information and resources brought by central board members will last after the IPO. In the specific context of emerging markets, these ties might be instrumental to secure privileged information and bail out from economic upheavals. Consequently:

Hypothesis 2: The higher the board centrality, the higher the IPO performance.

Board Experience and Long-Term IPO Performance

The presence of experienced directors on the board signals to outside investors that the issuing company has an efficient corporate governance framework that understands how the process through IPOs occurs (Gray & Nowland, 2013). Thus, it is expected that experienced boards closely monitor firms’ decision-making processes (Payne et al., 2009) and exercise effective vigilance on the management process (Kroll et al., 2008). The Brazilian context requires that board members be experienced in the specific burdens associated with the IPO process. Conversely, non-experienced investors and board members might fail to identify subtle contractual elements or regulatory aspects that might impact the firms’ governance (Folta & Janney, 2006). Although initially there is no direct empirical support for the effect of board experience on IPO performance, Kor and Sundaramurthy (2009) point out that board members with previous experience in founding companies tend to be linked to greater sales growth. Furthermore, Field and Mkrtchyan (2017) found that the higher the percentage of board members with previous experience with acquisitions, the greater the abnormal return measured through the CAR. More recently, Chahine et al. (2021) also identified a positive effect of experience on IPO performance, however, they looked at the effect of prior experience of venture capital funds. Thus, even though previous studies that directly analyzed the effect of the board's previous experience with an IPO on performance were not identified, we understand that the rationale is valid.

Thus, we propose that the presence of directors with previous IPO processes work as a credible signal either because those members help the company to implement the IPO process, or because the company has an efficient governance model that makes it possible to hire experienced board members, and:

Hypothesis 3: The greater the board experience with previous IPO processes, the higher the IPO performance.

The Tradeoff Between Private Equity Investment and Board Centrality and Experience

As seen before, association with Private Equity funds are perceived as a strong quality signal by the market. Yet, when considering investment opportunities, private equity firms face the risk of ‘adverse selection’, as they are uncertain whether the new portfolio member will perform well after the investment event (Akerlof, 1970). The source of this risk lies on the information asymmetry and uncertainties associated with the investment opportunity. On the one hand, the prospect firm's management possesses better understanding of the business. On the other hand, even when information asymmetry is decreased, investors still face the uncertainty on how an asset will perform under different circumstances (Groysberg & Lee, 2009). Furthermore, it is unclear to private equity investors the potential portfolio firm managers’ intentions and willingness to commit to future challenges (Bergh, Ketchen, Orlandi, Heugens, & Boyd, 2019). In order to address these sources of risk, private equity firms impose several obstacles to accept a firm in its portfolio. On the one hand, private equity firms require implementation of new managerial practices that are simultaneously perceived as improving quality facilitating transparency, thereby reducing information asymmetry.

Practices associated with accountability and transparency might be costly to implement (Arikan & Capron, 2010). As a consequence, they constitute an appropriate quality signal to potential private equity investors and ultimately to the stock market (Spence, 1978). Conversely, we posit that managers who perceive their firms’ quality as low or are not willing to commit to private equity's performance expectations might shy away from associating with private equity funds. Instead of associating with private equity firms, managers who perceive their firms as less competitive might attempt to choose for its board experienced and central directors in the Brazilian board interlock. By choosing central directors in the board interlock, these firms signal proximity to public agencies and governmental actors that might be instrumental as a safeguard against abrupt performance upheavals.

In other words, while we expect that firms attempting to signal high competitiveness self-select to be associated with private equity firms, we expect that firms less certain on their quality and commitment will attempt to be closer to potential rescuers. Consequently, in the absence of more prominent signaling, that is, the association with a Private Equity fund, firms will try to signal a better likelihood of being successful in the IPO process through other signals, such as the centrality and experience of the directors which were allocated on the board. Hence:

Hypothesis 4a: The positive effect of board centrality on IPO performance will be higher for companies without PE investment than for those with PE investment.

Hypothesis 4b: The positive effect of board experience on IPO performance will be higher for companies without PE investment than for those with PE investment.

METHODS

Data and Sample

In order to construct the database, we gathered information on all Brazilian IPOs from 2004 to 2013 that were issued in the B3 (formerly, BM&F Bovespa), the Brazilian stock exchange. In Brazil, information related to board composition and IPO prospectus is public. Signals of the quality of the firm going to IPO are especially important in emerging markets, which have lower property rights, institutional voids, and information asymmetry.

We began our evaluation series in 2004 because that year was marked by the resumption of private equity funds participation in the IPOs. We suspended collection in 2013 due to the absence of initial public offers materializing in the year 2014, as well as because of the Brazilian political–fiscal crisis that proceeded the year of 2015. However, some of the companies that issued stocks in the period closed the capital before the valuation period of the stock return. Others had liquidity problems. Finally, we removed four cases that presented as outliers, with a standard deviation greater than 3. Thus, our sample comprised 120 initial public offerings in the period, whose temporal distribution can be seen in Figure 1. There is a concentration of issues in 2007: 45% of the sample and, in that year, the mean leverage is also higher. Private equity-backed IPOs represent 56% of the sample.

Figure 1. Initial public offerings in Brazil between 2004 and 2013

The market and financial indicators of each company were obtained from the Economatica® databank. The other data on the company characteristics, board composition, and directors’ demographics and experience were manually collected from the system of the Brazilian Securities Commission (CVM). Until 2008, the CVM offered this information in its Annual Information Report (IAN). Between 2009 and 2013, data migrated to the Reference Form.

Variables

Table 1 shows the descriptive statistics and correlation matrix of the variables. In the sequence, we detail the operationalization of the variables.

Table 1. Descriptive statistics

Notes: N = 120; **p-value < 0.01; *p-value < 0.05.

Dependent variable

IPO's Cumulative Abnormal Return (CAR). In theory, an asset's price should reflect all available information to the stock market (Brown & Warner, 1980; Fama, 1991). New events feed the market's actors with new information, which triggers a reassessment of the asset's value (MacKinlay, 1997). For that reason, it is expected that the IPO event signal to the market is that a firm is going through several internal changes and will become more attractive for investors. If that signal is a credible signal, it should support expectations for positive performance.

In this study, we used the 1-year CAR to measure IPO performance and, consequently, the credibility of the signal. This measurement does not include the return component explained by the market, representing only the return as explained by specific characteristics of the IPO. Following previous studies (Brav, Geczy, & Gompers, 2000; Hwang, Titman, & Wang, 2018; Wang, Su, Coakley, & Shen, 2018), we adopted a 1-year horizon. Many IPO studies use underpricing (usually measured as the one-day CAR or simply as the return measured by the first day closing price to launching price) and other performance variables associated with the firm's valuation upon the IPO event. While we recognize the benefit of these popular metrics, we preferred to use one-year CAR due to five reasons: (1) Its previous usage in signaling and IPO literature (Bruton et al., 2010; Drebinger et al., 2019; Gao & Jain, 2011; Gibbs & Hao, 2018); (2) Underpricing fails to assess short-term performance (Park, Borah, & Kotha, 2016); (3) The signaling effect of PE funds tends to be durable and prominent over long periods of time (Arikan & Capron, 2010); (4) In several contexts, floating prices after the IPO event suffer several distortions, such as partial adjustment (Bradley & Jordan, 2002), in which underwriters proposedly establish an issuing price below the market value estimated during the book building process, in order to compensate institutional investors for disclosing information. Such distortions eventually fade out during the first year upon the IPO; (5) The investors in Brazil still need credible signs to assess the quality of the management after one-year of issuance. The vast majority of the IPOs in our sample were issued in the B3 New Market segment. Until 2017, the regulation of this segment imposed a two-stage lockup period: insiders could sell 40% of their shares only after six months of issuance, and 60% of the remainder shares after one year (Talans & Minardi, 2021). Therefore, one-year after issuance the information asymmetry problem persists. This is when PE funds and key board members are able to sell their equity stake and leave the company, and consequently the skills of the management in running a public company without such support will be tested.

In operational terms, adopting the index model as the market equilibrium, the component of the return not explained by the market (abnormal return) is calculated according to:

(1)$$\hskip8pcAR_{i, t} = R_{i, t}\ndash ( \alpha _i + \beta _i\ast RM, \;t) $$

In which, AR_i,t is the abnormal return of stock i in time t; Ri,t is the return of stock I in time t; α is alpha, the constant of the regression of asset i's return to the market portfolio's return; β is beta, coefficient of the regression of asset I's return to the market portfolio return; RM,t is the return of the market portfolio at time t. We used IBOVESPA (the B3 benchmark index) as a proxy for the market portfolio.

To obtain α and β, we were required to run a regression on asset i's return to the market portfolio's return in a window of time before the event. As we did not have stock price series of asset i before the IPO launching date, we were unable to estimate these coefficients. We used a simplified approach, adopting α = 0 and β = 1:

(2)$$\hskip9pcAR_{i, t} = R_{i, t}\ndash RM, \;t$$

The returns were calculated daily and compared with the market portfolio return: IBOVESPA. The daily abnormal returns were calculated for one year after the firms’ IPOs. The CAR is the sum of these daily returns since the first public offering for one year. CAR, the cumulative abnormal return, is shown in the sequence:

(3)$$\hskip9pcCAR = \sum\limits_{t = 0}^\infty {ARt} $$

In which, CAR_T  is the cumulative abnormal return at period T. In our case, it is one year; ARt is the abnormal return at day t, where 0 is the event day.

We collected the prices of the offered stocks in the Economatica® databank up to one year from the IPO inception, totaling 252 days of observations for each case. We also collected the IBOVESPA index for the same dates of the prices to calculate the abnormal returns (AR) of each stock. In the final sample, to estimate one-year CAR, we only included IPOs that had at least 190 trading days after the issuing date.

Independent variables

Private Equity Fund Presence (PE). We used a dummy variable to indicate whether the company at the time of the IPO had a private equity fund as a shareholder, based on information in the IPO prospectus. Other studies used the same strategy to indicate the participation of private equity funds in IPOs (Gibbs & Hao, 2018; Goktan & Muslu, 2018; Sletten, Ertimur, Sunder, & Weber, 2018), showing that it was adequate to signal the presence of these funds. Moreover, as we proposed in two hypotheses that a lack of funds positively moderates the effect of board centrality and the director's experience on IPO abnormal return, we created an inverted dummy: Private Equity Fund Absence (PE0), whose absence was defined as ‘1’ and the presence as ‘0’. We used this inversion to facilitate the interpretation of the coefficients of moderation models since the effect size is the same.

Board Centrality (degree). To build the board interlock network, we collected information, not only about companies that made the IPO in the period, but also concerning the boards of directors of all companies with shares traded in B3 (Brazilian stock exchange) between 2002 and 2013. For each year, we checked all CVM reports and forms, capturing the director's names, characteristics, personal data, and experience.

We located a total of 9,117 directors who were seated on the boards of 623 companies. Then, we used a recurrent strategy in the literature to constitute a 2-mode network that associated directors with the boards of companies (Davis & Mizruchi, 1999; Fracassi & Tate, 2012; Johnson, Schnatterly, & Hill, 2013; Mendes-da-Silva et al., 2008; Zona, Gomez-Mejia, & Withers, 2018). For this reason, the relationship between individuals was indirectly inferred from co-participation in the same board. Because board directors participate in more than one board, a network of boards and directors emerged, comprising the majority of public firms. For each year, if an individual was a member of a certain board, he/she received the value ‘1’, and ‘0’ if he/she was not present.

In order to estimate the board centrality of companies who made initial public offerings, we applied 2-mode network approaches (Borgatti, 2009; Everett & Borgatti, 2005). It is important to emphasize that an IPO firm might have many members on its board. However, not all of them sit at other boards. For that reason, only the members who sit on other boards contribute to a company's centrality degree. Thus, the board centrality of each company was given by the sum of adjacent companies that shared at least one director.

IPO Board Experience. To capture the board company's experience in the initial public offering, we considered the ratio between the number of directors who participated in previous IPOs and the size of the board. We chose this measurement of experience with the initial public offerings since it is directly related to the IPO process. Because some studies using more generic measurements of previous experiences tend to present more inconsistent results (e.g., Gray & Nowland, 2013; Kor & Sundaramurthy, 2009; Payne et al., 2009), we opted to evaluate the specific board experience with the IPO process. The research strategy of evaluating the previous experience with the object whose performance is analyzed was recurrent in research on acquisitions and alliances (Cho & Arthurs, 2018; Field & Mkrtchyan, 2017; McDonald et al., 2008; Peruffo et al., 2018). Although we do not identify studies evaluating previous experience in IPOs, we consider that the validity is the same as that of other studies that used specific experiences as a proxy.

Control variables

The Year of 2007 (Bubble year)

That year comprised 45% of the IPOs in the sample, with higher than usual leverage. The market was overoptimistic in this particular year, and investors may have paid more for IPOs and accepted less attractive issues, in a pattern similar to the 1999–2000 internet bubble. As the period matters for performance (Ritter & Welch, 2002), we have identified IPOs issued in 2007 with a dummy variable.

2008–2009 Financial Crisis

Those are the years of global financial turmoil. The assets issued in the crisis period likely had peculiar behavior if compared with assets issued in a regular environment (Didier, Love, & Martínez Pería, 2012). In order to control for the impact of the time in IPO performance (Ritter & Welch, 2002), we used a dummy variable to indicate IPOs issued during the years 2008 and 2009.

Issuance volume

This variable controls for the volume of shares traded in the IPO, calculated as the amount the value of the shares in Brazilian real offered in each issuance. The higher the volume, the higher the liquidity of the investor, which, in turn, may affect the return of the IPO (Ritter & Welch, 2002). Investors require a premium for illiquidity (Amihud & Mendelson, 1986). The final value is the natural logarithm of the issuance volume to equalize the coefficient of the variable.

Market capitalization

This is a variable that controls the size of the company, as measured by market capitalization. The variable is calculated by taking the number of outstanding shares at the date of the IPO, multiplied by the issuance price, totaling the market capitalization of the company at the time of the IPO inception. We use the natural logarithm of the firm's market capitalization to equalize the coefficient of the variable. Larger companies have lower risk compared with smaller firms (Fama & French, 1992) and should be more prepared to go public (Pagano et al., 1998).

Primary offering

This variable indicates the percentage of the total issuance volume that relates to the primary offering, that is, that will be invested in the company. Usually, part of the offer is secondary, transferring shares of previous owners to new investors. We expect that a higher percentage of primary offerings will have a positive relation to CAR since it signals potential growth for the company (Huyghebaert & Van Hulle, 2006).

Underwriter

This dummy variable indicates whether the underwriter is Pactual, UBS, UBS Pactual, or Credit Suisse. These banks alone, from 2004 to 2013, were the underwriters of 53 IPOs, meaning 36% of the IPOs issued in the last 10 years, and it is a proxy for underwriter reputation. There is evidence in the Brazilian market that reputable banks that underwrite IPOs are related to higher returns (Almeida, 2011), although more recent studies indicate that evidence of underwriters’ conflict of interest mitigated the effect on IPO performance (Almeida & Leal, 2015).

Underpricing

This involves the listing of an initial public offering below its market value. Underpricing was captured by finding the difference between the offer price and market closing value on the first trading day of the stock market. As a measurement of IPO efficiency, we use this variable because there is strong evidence that it is related to the long-term return of the emissions (Chambers & Dimson, 2009).

Market-to-book ratio

The market-to-book ratio indicates the market value in comparison to the accounting value of the company, with the former being determined by the price in the stock market multiplied by the number of outstanding shares. The book value is the total equity value reported in the balance sheet of the companies as of the last quarterly press release available before the IPO date, e.g., if the IPO date is May 5, 2012, the balance sheet information is related to March 31, 2012. It serves as a proxy for prospects of the company growing (Pagano et al., 1998) and, according to Fama and French (1992), investors require a premium for the lower market-to-book ratio.

Financial leverage

This variable controls for a firm's leverage, calculated as total assets divided by total equity. Higher leverage increases the risk of shareholders (Brealey, Myers, Allen, & Mohanty, 2012). The total assets and total equity values taken were reported in the balance sheets of the companies as of the last quarterly press release available before the IPO date.

Models and Robustness Check

To test the hypotheses, we estimated hierarchical regression using the OLS models, with the long-term cumulative abnormal return (CAR), a proxy of IPO performance, as the dependent variable. These models are often used in studies that evaluate the performance of IPO (Chahine et al., 2021), or any other phenomenon whose past events are unknown or inaccessible (Field & Mkrtchyan, 2017). Thus, six models were estimated (Table 2), in which we used the robust standard error to deal with heteroscedasticity problems (White, 1980). In Model 1, we estimated the regression only with the control variables. Models 2, 3, and 5 were estimated in terms of each of the three independent variables, aiming to test Hypotheses 1–3. In Models 4 and 6, we tested Hypotheses 4a and 4b, moderating the board centrality and board experience variables by the dummy variable that represents the absence of IPO backed by private equity funds (PE Absence). As the absence or presence of participation of PE funds refers to a contextual condition, we tested the interaction variables in different models to avoid problems of collinearity, since focusing on the original variables would not reduce this problem (Jaccard & Turrisi, 2003). To statistically test for endogeneity, we checked through the Durbin-Wu-Hausman test if the OLS regressors provide the best linear unbiased estimator for our models. All tests pointed to the exogeneity of the estimators (p > 0.1), indicating that the OLS models were the most adequate. Finally, as the tradeoff hypothesis assumes that PE investment is the most relevant signal, we compared the effect size of the PE coefficients with those of the other independent variables using the F-test.

Table 2. Effect of variables on the IPO's cumulative abnormal return (OLS models)

Notes: Standard errors are in parenthesis. N = 120.

^a Delta R-squared calculated regarding Model 1, composed of control variables.

We checked the robustness of the results in nine ways. First, we evaluated whether the dependent variable presented a serious distribution problem. Second, we analyzed the functional form of the effect of each of the independent and control variables on the dependent variable. Third, via the White test, we checked to determine if the models presented some problem of heteroscedasticity. Fourth, we identified any outliers present to correct leverage problems and poor model specification. Fifth, we evaluated whether the variables had some collinearity problem through the evaluation of tolerance indicator and VIF (tolerance less than 0.2 and VIF greater than 5). Sixth, the presence of some unobserved effect (e.g., temporal disturbance) was inspected using the Durbin-Watson error autocorrelation test, which was within the acceptance limit. Seventh, in addition to assessing the significance of the coefficients, we checked effect size using the Eta-squared statistic to ensure that our interpretations of which variables had greater magnitude were correct. Additionally, we compared all models with the base model – composed only of the control variables – using the F-test of the R-squared variation (ΔR²), to make sure that the inclusion of the independent variables significantly increased the explanatory power. Eighth, industry rivalry can affect the IPO return of companies (Hou & Robinson, 2006), we verified the effect of the industry concentration through the Herfindahl-Hirschman index (HHI), which was neither significant nor changed the significance of the other variables. Finally, we verified whether the signaling effect varied across industries using mixed multilevel models, which were found to be non-significant. Descriptive statistics by industry are shown in Appendix Table A1.

RESULTS

Table 2 shows the results for models using the OLS method. We also report effect size of the independent variables using the eta-square (η ²) to complement the interpretation of the results, to demonstrate which of them has the highest coefficient of explanation in the models. Model 1 shows the effect of control variables on the cumulative abnormal return (CAR). The results indicate that IPOs carried out in the year of 2007 were negatively affected (β = −0.609, p < 0.001). Companies with higher underpricing had greater cumulative abnormal return (β = 0.942, p = 0.010). The degree of financial leverage was also significantly associated with higher CAR (β = 0.003, p = 0.080). In Model 2, the results indicate that the presence of private equity funds increases the firms’ CAR (β = 0.173, p = 0.015, η ² = 0.053), corroborating Hypothesis 1. This effect is economically significant because it represents a 17.3% cumulative abnormal return for private equity-backed companies. In addition, it corresponds to a model explanation coefficient of 5.3%. In Model 3, board centrality did not present significant effect on CAR (β = −0.001, p = 0.988, η ² < 0.001), leading us to reject Hypothesis 2. In Model 5, there was also no significant effect of the IPO Board Experience on the abnormal return (β = 0.144, p = 0.397, η ² = 0.006), refuting Hypothesis 3.

In Model 4, we test if the effect of board centrality on firms’ IPO performance by CAR will be higher for companies without PE investment than PE-backed companies. The interaction coefficient of PE Absence × Board Centrality indicates a positive and significant effect (β = 0.028, p = 0.047, η ² = 0.036), corroborating Hypothesis 4a. Among PE-backed companies, the effect of board centrality was moderately significant (β = −0.016, p = 0.075, η ² = 0.026). However, the effect was negative, reducing value of the CAR. The results show that for each additional tie within companies without private equity-backed, the CAR increases by 4.4% when compared with PE-backed companies (β _{PE Absence × Board Centrality} = 0.028 – β _{Board Centrality} = −0.016 = 0.044).

Still in Model 4, it should also be noted that the effect of PE-backed companies on CAR remained positive and significant, considering the inversion of the Private Equity Fund Absence variable (β _PE0 = −0.314, p = 0.001, η ² = 0.089). Furthermore, PE investment shows an effect significantly greater on CAR than board centrality, both among companies without PE investment (η ²_PE0 = 0.089 > η ²_{PE Absence × Board Centrality} = 0.036, F = 10.19, p = 0.002) and among PE-backed companies (η ²_PE0 = 0.089 > η ²_{Board Centrality} = 0.026, F = 10.45, p = 0.002). This result supports the assumption of a tradeoff between PE funds and other signals since the first is the strongest signal. In addition, inclusion of the moderator term of PE-backed companies in Model 4 increased the effect size of the board centrality when compared with Model 3, which reinforces the idea of a tradeoff between PE and the other signals.

In Model 6, we check if the effect of board IPO experience on CAR will be higher for companies without PE investment than for those with PE investment. The interaction coefficient of PE Absence × IPO Board Experience (%) also points to a positive and significant effect at 90% (β = 0.467, p = 0.094, η ² = 0.020), corroborating Hypothesis 4b. However, the effect of IPO board experience on CAR was not significant among PE-backed companies (β = −0.210, p = 0.326, η ² = 0.006). The difference between the coefficients of variable IPO Board Experience (%) of private equity-backed and not-backed companies is 0.677 (β _{PE Absence × IPO Board Experience (%)} = 0.467 – β _{IPO Board Experience (%)} = −0.210). Thus, for a 10% increase in the number of board members with previous IPO experience, the results point to a 6.77% increase in CAR among PE not-backed companies to PE-backed companies.

We also checked in Model 6 whether the effect of PE-backed companies on CAR remained the strongest signal. The effect size was 6.6% on CAR (β _PE0 = −0.258, p = 0.006, η ² = 0.066), greater than the effect of board experience, both among companies without PE investment (η ²_PE0 = 0.066 > η ²_{PE Absence × IPO Board Experience} = 0.020, F = 3.63, p = 0.059) and among PE-backed companies (η ²_PE0 = 0.066 > η ²_{IPO Board Experience} = 0.006, F = 1.62, p = 0.20), in which the last test was not significant because the IPO Board Experience variable (%) was no longer significant, affecting the value of the F-test. Again, the results demonstrate that the presence of PE funds is the most relevant signal, whose tradeoff is also manifested in signaling through more experienced boards. However, unlike what happened regarding board centrality, the inclusion of the moderator term of PE-backed companies in Model 6 increased the effect size of the board experience when compared with Model 5 only for companies without PE investment. Which does not invalidate Hypothesis 4b but leads to additional analyses.

Post-Hoc Exploratory Data Analysis

We illustrated the results of Models 4 and 6, which refer to tradeoff Hypotheses 4a and 4b in Figure 2. The initial intuition of our hypotheses was that the effect of board centrality was accentuated by the absence of private equity funds. However, as can be seen in the upper graph, there are signs of a reverse effect, in which the board centrality increases the CAR in the absence of PE funds and reduces the CAR in their presence. As the independent variable and interaction term have a different directionality, and both are significant, they point out that this reverse effect may have a meaningfully crossing point (Gardner, Harris, Li, Kirkman, & Mathieu, 2017). Dividing the negative value of the coefficient of moderating variable by the interaction term (X _cross = −β _{PE Absence} = −0.314/β _{PE Absence × Board Centrality} = 0.028), our data show that the reversal of board centrality on the CAR occurs when there are more than 11.21 ties in a company: above this value, the effect of board centrality is not sufficient to reverse the effect of CAR among PE not-backed companies.

Figure 2. Moderation of private equity fund presence on board centrality (degree) and board IPO experience effects Legend: IPO's Cumulative Abnormal Return (CAR); Private Equity Fund Presence on Initial Investment (PE); Board Centrality (Degree).

The graph below shows a similar relationship but in terms of the previous experience of the board with IPO processes. IPO Board Experience and the interaction term PE Equity Fund Absence have a different directionality, however, only the second is significant. The graph points to a reverse effect, but as one of the regions (PE-backed) is not significant, the effect is more likely to be a substitution effect (Gardner et al., 2017). Thus, the results show that, among PE not-backed companies, boards that have more than 55.2% of directors with previous experience in IPO processes (X _cross = −β _{PE Absence} = −0.258/β _{PE Absence × IPO Board Experience (%)} = 0.467), tend to have a positive effect on the CAR. Among PE-backed companies, the negative effect was not significant.

DISCUSSION

Throughout this article, our goal was to explore how different quality signals impact IPO valuation. Within the context of emerging economies, we claimed that board centrality and experience not only signaled quality but also survival likelihood. Within underdeveloped and turbulent markets, influential and experienced board members might be able to facilitate access to scarce resources, frequently from public and State agencies and economic groups.

Brazil offers a specific case among emerging markets. Brazil is considered a ‘State as minority investor’ type of capitalism where, despite the reduced role of the State in the economy, close relationships to public and state agencies are important (Musacchio et al., 2015). In that context, these relationships are frequently associated with access to political influence that might directly or indirectly favor the firm. At the same time, many private equity firms were instrumental in supporting IPOs, leading several IPO firms to adopt demanding managerial practices. The contrast between these different alternatives led us to test them simultaneously.

Within this context, being qualified to attract Private Equity investment constitutes a crossroad for IPO firms. The managerial practices and changes required by Private Equity firms prove to be a burden to several firms, who might be unable to undergo these adjustments. Conversely, firms that are unable to attract private equity firms might pursue central and experienced board members. Thus, this article's major contribution to literature lies in the tradeoff firms face between attracting Private Equity investment or relying on experienced and central board members. Our evidence shows that the effect of director experience and board centrality are mechanisms contingent on the presence of PE funds; the centrality of board members in the interlock community is only conducive to higher abnormal gains when firms are not associated with private equity firms. Also, the market only recognizes the importance of directors with more previous experience in IPO when PE funds are absent.

Contributions to the Literature

Role of private equity funds on IPO performance

Scholarly research has established the association between PE funding and IPO performance (Bruton et al., 2010; Lee et al., 2011). This result is consistent with PE funds promoting other value creation actions in their invested companies that increase the odds of successful IPOs, or PE investing in companies with the best governance and managerial practices due to the meticulous selection process. These findings resonate with the insight that being associated with a PE fund is a costly signal, not attainable to all IPO firms. Yet, our post-hoc analyses permit us to highlight unexpected dynamics, not originally covered by our hypotheses: the greater the board centrality, the lower the IPO performance among PE-backed companies.

Within the Brazilian context, when a PE-backed firm engages in bringing in board members that are central in the board interlock, it might send a dubious signal to potential investors. Black, De Carvalho, and Gorga (2010) highlight that companies in emerging markets that experience poor results or have some disputed practice seek to signal changes in corporate governance by more central board members. Thus, reliance on central board members might signal to the market that the firm might require further assurance through political ties, which signals low quality to potential investors. Thus, these findings lead us to suggest that the literature on private equity investment should take into account that positive signals originated from PE funding might be offset by negative signals stemming from bringing in central board members. We posit that this finding is generalizable to economies where state intervention is likely to take place (Wright et al., 2021).

Role of board centrality on IPO performance

Extant literature has amassed evidence that board centrality leads to higher IPO performance (Filatotchev et al., 2018; Reuer et al., 2012). This relationship is expected to be more important within emerging economies, as the market failures associated with institutional voids and underdeveloped capital markets are frequently offset with board members’ relationships and prestige. We expected that the Brazilian case would reinforce this association since board members direct and indirect ties to State and public agencies and economic groups are important sources of competitive advantage, and their state intervention is likely to take place (Wright et al., 2021). Although we found this association, this effect was not always important. The contingency effect of board centrality suggests that board members’ relationships are not always relevant. This finding contrasts with the prevalent understanding that board centrality is regularly associated with higher performance. Only firms that are unable to espouse private equity's requirements to receive the latter's investment will rely on board centrality as the main quality signal.

Role of board experience on IPO performance

Extant literature associates board experience with IPO performance (Cho & Arthurs, 2018; Field & Mkrtchyan, 2017). Analogously to the reasoning employed in understanding board centrality, we expected that board experience would also be a prevalent explanation, for experienced board members would have accumulated relevant local and specific knowledge to advise on IPO processes in Brazil. Yet, we found that the effect of director experience is contingent on the presence of PE funds; the market only recognizes the importance of directors with more previous experience in IPO when PE funds are absent. Our study used prior directors’ experience in IPO processes as a proxy for board experience and expertise. To the extent of our knowledge, this is a pioneer usage, thereby introducing a methodological contribution. Other studies have used specific measurements of experience (Cho & Arthurs, 2018; Field & Mkrtchyan, 2017; McDonald et al., 2008; Peruffo et al., 2018), but did not consider experience with the IPO process itself. In our study, we demonstrate how recurrent experiences in the same domain can leverage knowledge and capacity which, in turn, can promote greater performance.

Limitations and Suggestions for Future Research

Future research should assess the generalizability of these findings to other contexts. Brazil is characterized as a ‘State as a minority investor’ type of capitalism, where there is considerable threat of state intervention (Musacchio et al., 2015; Wright et al., 2021). In other emerging markets where the threat of state intervention is less likely (e.g., Singapore), ties to government officials and politicians might present lower influence on IPO performance. Firms without Private Equity investment might observe lower benefits as they associate with central board directors, for there are lower chances that the State will bail out failing companies. Conversely, the benefits associated with private equity firm investments should be less affected by the presence of ties to central interlock directors, as the latter will represent a weaker signal of quality.

Throughout this research, we have assumed that players aim at maximizing IPO performance. In contrast, alternative lenses might emphasize the role of power and dependency on decision-making. For instance, Adams, Hermalin, and Weisbach (2010) point out that board members appointed by PE funds tend to have greater bargaining power, reducing the power of external members, the CEO, and other members. As a result, some companies may try to compensate for this power loss by incorporating more external or experienced members and shield the board against PE funds’ influence. This conflict of interest becomes stronger when Private Equity investors and the firms’ management present distinct risk-taking behavior (Gillan & Starks, 2000). Future research might aim at disentangling these distinct mechanisms and identify why PE-backed firms that hire central board members hurt their IPO performance.

CONCLUSION

In this article, we investigate how IPO companies choose the most efficient signals to increase the odds of a positive IPO abnormal return. Our results indicated that the participation of private equity funds is the most relevant signal, while board centrality and board previous experience with IPO processes positively affect the IPO abnormal return only in cases of companies that are not backed by private equity funds. We also found a reverse effect of the board centrality on the IPO's performance: while higher board centrality increases the IPO abnormal return among companies not backed by PE funds, among PE-backed companies the effect is negative. The effect of the board's previous experience with IPO processes, on the other hand, points to a substitution effect: at higher levels, it replaces the absence of private equity fund coverage, having no significant influence among PE-backed companies.

DATA AVAILABILITY STATEMENT

Data required for replicating these results are available at: Kirschbaum, Charles; Rossoni, Luciano; Minardi, Andrea; Borges da Silva, Emília (2022). Data for ‘The Tradeoff between Private Equity Sponsorship, Board Centrality, and Experience as Credible Signals for IPO Performance’, Management and Organization Review. Mendeley Data. https://doi.org/10.17632/FC3HDZX5RF.1