Chairperson and CEO foreignness and CG quality of emerging markets MNCs: Moderating role of international board interlocks

We examine whether foreign chief executive officers (FCEOs) and foreign independent board chairpersons (FIBCs) improve on the corporate governance (CG) practices of emerging market multinational corporations (EMMNCs) through governance spill over. We use hand-collected data for 80 listed Nigerian multinational corporations for the period 2011 – 2016 (480 firm-years) and apply a three-stage least squares regression to address endogeneity issues. Our findings show international exposure of EMMNCs motivate appointment of FIBCs and FCEOs who positively affect their CG quality. In addition, international board interlocks positively moderate the likelihood of FCEOs to export and enhance EMMNCs' CG quality, but negatively moderate FIBCs impact on CG practices of EMMNCs. Finally, we develop a framework to show how EMMNCs' CG practices are exemplary to local firms in the home country who may mimic these governance practices. We contend the repeated game of governance spill-over and mimetic isomorphism drives the evolution of CG institutions and, potentially, will generate institutional change in CG practices in emerging markets.

This propagation of CG codes in emerging markets has coincided with the emergence of emerging market multinational corporations (EMMNCs) with significant operations in both developed and developing countries. Several studies have explored the impact of the unique institutional context of emerging economies on firm's outcomes (Areneke, Yusuf, & Kimani, 2019;Finchelstein, 2017;Machokoto, Areneke, & Nyangara, 2020;Sarhan, Ntim, & Al-Najjar, 2019;Tunyi, Agyei-Boapeah, Areneke, & Agyemang, 2019;Wang, Hong, Kafouros, & Wright, 2012;Yaprak, Yosun, & Cetindamar, 2018). The consensus within this literature is that, emerging markets are characterized by poor quality regulations, weak regulatory enforcement, political instability and the presence of corruptionfactors that limit the effectiveness of CG regulation within this environment and, potentially, impact on the ability of EMMNCs to compete on the global stage (Bhaumik, Driffield, Gaur, Mickiewicz, & Vaaler, 2019;. Similarly, recent studies have explored how multinational firms mitigate and manage institutional tensions between host and home countries through CG mobility or spillover (Aggarwal, Erel, Ferreira, & Matos, 2011;Miletkov, Poulsen, & Wintoki, 2017). For example, Aggarwal et al. (2011) find that foreign ownership acts as a channel for the transfer of good CG practices from one country to another. Similarly, Miletkov et al. (2017) provide evidence that foreign directors can export good governance practices into weak institutional environments. Despite this advancement in the literature, there is a dearth of studies exploring the role of EMMNCs and their leadership in promoting (hindering) the development of CG practices in emerging economies. We contend, due to coercion from abroad, EMMNCs and their leadership may play a vital role in enabling the propagation of good CG practices and strengthening the development of strong regulatory enforcement mechanisms in emerging markets. We address this research gap by examining how the internationalization of firm leadership (CEO and board chairman positions) contributes to the evolution and development of resilient governance institutions in EMs through diffusion of good CG practices from abroad into emerging economies. In addition, we examine the role of international board interlocks in shaping the evolution of CG practices in firms and boardrooms led by non-native directors.
International business literature has shown that compared to single country firms, internationalization coerces multinational firms to adopt institutional isomorphic strategies that enable spillover and adoption of similar practices across countries Bhaumik et al., 2019;Fainshmidt, Judge, Aguilera, & Smith, 2018;Hooghiemstra, 2012). We draw on this literature (institutional isomorphism) and repeated game theory to uncover how foreign chief executive officers (FCEOs) and foreign independent board chairpersons (FIBCs) improve on the governance practices of EMMNCs in their home country through CG transfers from abroad.
Following Cumming et al. (2017), we operationalize CG mobility and isomorphism as the diffusion or transfer of good CG practices from one country to another. Specifically, normative pressures from the international operation of EMMNCs coerces them to adopt good CG practices in an effort to achieve legitimacy and competitiveness, and to reduce liability of foreignness in their operations abroad. The pressure to adopt good CG practices to ensure legitimacy and reduce liability of foreignness may coerce EMMNCs to appoint foreign leaders to chair boardrooms (FIBCs) and manage firm operations (FCEOs). This appointments may also enable EMMNCs to attract foreign investors who may otherwise be discouraged by the challenging institutional environment and negative informal practices in EMs (e.g., corruption).
Foreign leaders (FCEOs and FIBCs), potentially, bring independence, human capital, experience and exposure to different governance institutions. They may, therefore, overcome the institutional void in EMs by reinforcing compliance with normative (voluntary) CG guidelines in the EMMNCs home country, which improves the quality of their governance practices. Furthermore, we conjecture that the repeated game of transfers through compliance with CG practices as recommended by regulators in the EMMNCs home country improves on their governance quality. The governance isomorphism process, curbs weak governance enforcement in EMs while reducing the liability of foreignness and improvement in cognitive legitimacy in foreign host countries. This arguably gives EMMNCs competitive advantage over their peers, especially at home. Consequently, due to the high quality of EMMNCs' governance practices, they may be more successful in the home country and a source of emulation by peer firms. We contend, emulation of EMMNCs' governance practices by local firms (mimetic isomorphism) in a repeated game may evolve to more resilient governance institutions at macro level which may bring about institutional change in CG practices in EMs.
We test our conjecture using mostly hand-collected data for 80 listed Nigerian MNCs for the period 2011-2016 (480 firm-years). We use compliance with the Nigerian Security and Exchange Commission (SEC) code of good practices in CG as a measure of the quality of MNC CG practices. To control for endogeneity, we adopt a threestage least square (3SLS) regression approach. Our results show a significant positive effect of foreign CEOs and FIBCs on the quality of governance practices of EMMNCs. Furthermore, while international board interlocks positively moderate the ability of FCEOs to transfer and promote good governance practices in firms, it negatively moderates the likelihood of FIBCs to export and enhance the governance practices of EMMNCs. Drawing on these findings, we make several contributions to CG literature.
First, we extend institutional isomorphism and repeated game literature by developing a conceptual framework ( Figure 1) showing how EMMNCs contribute to institutional evolution and change by using their internationalization of leadership roles to improve on the quality of CG practices at home. We show that coercion from abroad enables EMMNCs to continuously improve on CG practices in line with normative guidelines in the home country, which may lead to the development of more resilient governance structures at the firm-level over time (micro level change). The reputation of EMMNCs as exemplification of quality CG practices may result to mimetic isomorphism as peer firms in the home country imitate their CG practices. Through a repeated game of mimetic isomorphism, governance practices of firms in EMs evolve. Thus, leading to the development of more resilient governance institutions at the macro level, which can overcome weak regulatory enforcement (institutional void) at home and may generate institutional change over time.
Second, we extend the growing literature on CG mobility. As highlighted earlier, there is growing literature on the channels through which CG practices are transferred across economic institutions. However, these studies have concentrated on foreign directors (as a percentage of board composition; Dauth, Pronobis, & Schmid, 2017), cross-listing (e.g., Temouri, Driffield, & Bhaumik, 2016) and foreign ownership (Aggarwal et al., 2011) as mechanisms of governance mobility. Relatively scant attention has been given to the role foreign leadership of EMMNCs play in governance mobility despite the fact that, the positions of CEO and chairperson have been argued to have powerful influences on the direction and governance of firms (Clark, Murphy, & Singer, 2014;Coles & Hesterly, 2000). We address this gap by showing that FCEO's and FIBCs are important strategic nodes of isomorphism between home and host countries CG practices. Specifically, when EMMNCs employ FCEOs and FIBCs, these individuals bring diverse experience from their international exposure, socio-economic, institutional, cultural, business, technical and professional backgrounds in the leadership of EMMNCs. These attributes provide FCEOs and FIBCs with the necessary skills and knowledge to impact on the quality of CG practices in the home country of EMMNCs through exportation of good governance practices from abroad.
Finally, we highlight the international perspective of board interlock. The existing research on board interlocks mainly explores how national interlocks affects governance behaviour of firms albeit with mixed evidence (e.g., Fich & White, 2005;Pombo & Gutiérrez, 2011). However, to our knowledge, there has been no attempt to examine whether directors who seat on foreign boards (international interlocks) may impact (moderate) on the ability of agents of governance mobility to transfer F I G U R E 1 The conceptualization of how emerging market multinational corporations governance quality generates evolution of governance institutions in emerging markets governance practices across countries. We address this research gap by showing that international board interlock's moderate the ability of FCEOs and FIBCs to export and improve governance practices of EMMNCs at home. We show that, when foreign board interlock's increases, the international exposures of the board enhance FCEOs ability to influence EMMNCs' CG practices positively. On the other hand, increase in international interlocks can lead to director busyness, which may limit their meeting attendance, and involvement in CG practices of boards. This may negatively affect the foreign chairpersons' ability to influence EMMNCs' governance practices as they may lack support during board decisions because of absentee directors (including the foreign chairperson who may be limited by both geographical and outside board commitments) which reinforces the ability of CEOs to influence the quality of EMMNCs' governance practices.
The remainder of the article is structured as follows. The next section (Section 2), discusses the theoretical and conceptual framework of the study and testable hypotheses are developed. In Section 3, we present the peculiarity of the research context (Nigeria) and discussions of research design. Section 4 present and discuss the findings of the research. Finally, in Section 5, we summarize and conclude.

| LITERATURE REVIEW AND HYPOTHESIS DEVELOPMENT
Before discussing our main theoretical positions in this research (i.e., institutional isomorphism and repeated game perspectives), we note however that, the wider CG literature stems from a host of complementary theoretical standpoints (see., Gaur, Kumar, & Singh, 2014;Kumar & Zattoni, 2015). As such, we invoke such complementary theoretical lenses including resource dependency, agency and human capital theories to provide a comprehensive articulation of our research hypothesis.

| Institutional isomorphism and repeated game perspectives
Institutional isomorphism as a strand of institutional theory has received considerable interest in CG scholarship in recent years (Boxenbaum & Jonsson, 2017). Institutional isomorphism theorizes that there are powerful and diverse institutional influences on firm behaviour (DiMaggio & Powell, 1983). These institutional forces can promote or constrain certain activities of the firm. The consequence of these institutional forces push firms to adopt similar behaviours within and across economic environments (Cumming et al., 2017;DiMaggio & Powell, 1983;Fainshmidt et al., 2018;Gaur et al., 2014). The similarity in firm behaviour can arise because of coercion (coercive isomorphism) from formal institutions (e.g., code of practices of CG instituted by regulators) or informal traditions and norms. The other two forms of similarity arise due to pressure from institutionalized professional bodies (normative isomorphism) and mimetic isomorphism through imitation of the behaviour of more successful peer firms (DiMaggio & Powell, 1983;Gabbioneta, Greenwood, Mazzola, & Minoja, 2013;Greenwood, Raynard, Kodeih, Micelotta, & Lounsbury, 2011).
Repeated game theorizes that the behaviour of economic agents (players) occurs repeatedly over time (Atakan & Ekmekci, 2012;Bohnet & Huck, 2004;Mailath et al., 2006;Yoon, Guffey, & Kijewski, 1993). The repeated interaction between economic agents enables the likelihood of reputation effects whereas dynamic punishments sustains co-operative behaviours (Bohnet & Huck, 2004;Mailath et al., 2006;Yoon et al., 1993). For example, with regards to CG practices of firms, one of the players are regulators who institute formal governance laws which firms need to adopt annually (stage game) and which over time (repeated game) improves the firm's reputation. On the other hand, are stakeholders (players) who seek private benefits from firms by encouraging negative informal practices that limits transparency and accountability. The latter practices may be detrimental to firm's adoption of good CG practices over time (repeated game) but are damaging to the firm's reputation. Specifically, emerging market firms operate in home countries with formal and informal unwritten codes of practices that powerfully impact on their behaviours (Baker, Gibbons, & Murphy, 2002). For example, there are often informal practices (e.g., corruption) and unwritten codes (e.g., bribery) between firms and the external environment.
Consequently, both formal governance players (e.g., CG regulators) and informal governance players (e.g., corrupt officials and elites) affect the choice of the firm's governance practices. As a result, firm-level governance practices reflect the interplay of repeated games between formal and informal governance practices. The formal governance regulations encourage firms to adopt good governance practices, which improve governance quality and attract investors. However, adoption of these practices is costly, and weak enforcement further disincentivises their adoption. On the other hand, strong informal practices such as corruption have a detrimental and adverse reputational effect on firm governance practices, but are cheaper alternatives to access critical resources especially in EMs where elites and politicians are private benefit maximizers and gate keepers to resources. Furthermore, weak enforcement of formal governance guidelines may reinforce negative informal practices such as bribery and corruption. For example,  show institutional void enforces the ability of elites to invent, circumvent and corrupt institutions which work against formal regulatory initiatives to improve governance practices in EMs. However, informal governance practices such as corruption and elitism expose firms to economic, political, social and environmental cost. Therefore, given the costs and benefits, the optimal CG practices of firms in emerging economies can either maintain or change governance institutions over time depending on the tradeoff between formal and unethical informal governance practices.
Drawing on both institutional isomorphism and repeated game theorizing, we argue that the internationalization of EMMNCs poses as external coercion and influence (added player) on firm governance practices compared to firms operating only in one country. The international exposure of EMMNCs coerces them to adopt governance isomorphism strategies that enables diffusion of good governance practices from abroad to improve on the home country CG practices. Furthermore, through governance isomorphism, EMMNCs improve on their cognitive legitimacy abroad; reduce the liability of foreignness, cultural distance, governance cost and institutional void at home while simultaneously attracting foreign and local capital. Governance isomorphism reinforces the adoption of recommended formal CG guidelines in the home country while limiting the problem of weak enforcement by regulators. On the other hand, bad governance also travels internationally (Allred, Findley, Nielson, & Sharman, 2017;Cumming et al., 2017). For example, EMMNCs can also be attracted to countries with poor governance and weak regulatory oversight to benefit from corrupt opportunities.
However, we argue that on aggregate, EMMNCs will trade off unethical informal practices for formal governance practices that ensure accountability and transparency, which improves both local and international reputation of the firm. The repeated game of governance isomorphism by EMMNCs evolves to robust, resilient and quality CG practices and makes the governance practices of these firms reputable and attractive to peer firms in the home country. This leads to mimetic isomorphism as peer firms imitate the CG practices of EMMNCs seen as more successful. The repeated game of mimetic isomorphism of governance practices leads to the evolution of resilient governance institutions that may evolve to overcome institutional void and generate institutional change in EMs. Figure 1 shows our conceptual framework, which uncovers how EMMNCs bring institutional evolution in governance practices in EMs through their leadership. Specifically, internationalization of EMMNCs coerces them to adopt isomorphic governance practices across economic environments. This motivates EMMNCs to appoint FCEOs and FIBCs from host countries especially those from countries with robust CG guidelines and enforcement. Foreign independent board chairpersons (H1) and Foreign CEOs (H2) bring their international experience, background, knowledge and skills, which enables them to improve the governance practices of EMMNCs through governance isomorphism. Isomorphism of governance practices through FCEOs and FIBCs is moderated by the level of directors interlock/ affiliation with boardrooms in other countries (H3 and H4). The repeated game of improvement of EMMNCs' CG practices through governance isomorphism (H1-H4), generates more resilient CG practices at home. Mimicking of EMMNCs' CG practices (right-hand side of the figure) by peer firms leads to improvement in the latter's CG practices. The repeated game of imitation by peer firms and governance isomorphism by EMMNCs evolves to macro-level institutional change in governance practices in EMs (see extreme right of Figure 1). Drawing on the conceptual framework (Figure 1), we develop four testable hypotheses.

| Chairperson foreignness and the quality of CG practices
Board leadership debate is generally discussed across two streams of literature. On the one hand, drawing from agency theory, research on board leadership independence argues that the separation of chairman and CEO positions leads to a balance of power which limits insider dominance (e.g., see Baliga, Moyer, & Rao, 1996;Kula, 2005;Nicholson & Kiel, 2007;Pearce & Zahra, 1991). These studies mainly argue that the separation of the positions of CEO and chairperson is a more effective measure of monitoring and control. Specifically, when there is CEO-chairman duality, the board's ability to monitor and control management decreases that, consequently, leads to a lack of independence and agency conflict. This negatively affects firm CG practices and shareholders' value maximization.
On the other hand, organisational leadership and strategic management literature offer insights on how the backgrounds, skills', leadership styles and experience of CEOs and board chairpersons can affect firm outcomes.
Specifically, this literature argues that the presence or absence of certain leadership characteristics influences organizational direction, governance and performance (Fitzsimmons & Callan, 2016;Puffer & Weintrop, 1995;Tian, Haleblian, & Rajagopalan, 2011). For example, Clark et al. (2014) contend that different governance and ownership structures enhance (or limit) the ability of corporate leaders to improve financial performance. Similarly, Fitzsimmons and Callan (2016) find that CEOs' social capital and industry-specific knowledge are traits sought after by boards when seeking to appoint new CEOs.
Existing studies have, however, overlooked how the interface between leadership independence and leadership human capital integrates with the international perspective of firms in promoting institutional isomorphism practices. We attempt to close this lacuna with insights from international business research. Specifically, recent international business research has shown that in the process of governance mobility, MNCs can impact on home country governance practices through international governance agents including board of directors with different nationalities, foreign ownership and crosslisting (e.g., see Cumming et al., 2017;Dauth et al., 2017;Doh, Rodrigues, Saka-Helmhout, & Makhija, 2017;Hooghiemstra, Hermes, Oxelheim, & Randøy, 2019;Miletkov et al., 2017). For example, Miletkov et al. (2017) report that multinational directors improve the governance disclosure quality of firms. Similarly, Temouri et al. (2016) show that dual listing improves firm-level governance quality through bonding. Despite these recent contributions, whether the foreignness of EMMNCs leadership improve on their CG quality at home remains a theoretical and empirical gap.
We use insights from leadership independence, human capital and governance mobility literature to close this research gap by examining how foreign leadership of firms through the positions of CEOs and chairpersons enhance EMMNCs' governance quality at home. Specifically, we explore the interface between the three streams of literature by showing that the internationalisation of EMMNCs encourages them to develop institutional isomorphism practices across economic institutions. This is particularly relevant in EMs where isomorphism can reduce liability of foreignness and improve cognitive legitimacy abroad while overcoming the weaknesses in the home country especially institutional void. Specifically, compared to local firms, internationalization of EMMNCs pushes them to appoint FCEOs and FIBCs to lead the firm. These foreigners bring with them diverse human capital due to their heterogeneous origins, which they employ to enhance the quality of CG practices in the home country.
We argue that independent foreign chairpersons occupy a strategic position, which enables them to influence boardroom behavior of EMMNCs in adopting good CG practices. Extant literature shows the separation of CEO and chairman positions enhance the firm's governance practices (Kula, 2005;Nicholson & Kiel, 2007). In this light, many EMs have developed codes that emphasize separation of these leadership positions to ensure board independence. For example, the Nigerian Securities and Exchange Commission (2011) prescribes the separation of leadership position. It is therefore unsurprising that most Nigerian firms (presented later) have separated their leadership positions.
We contend that foreign leadership of a separated CEO and chairperson position in EMs offer additional perspective on leadership independence. This is because foreigners are less likely to collude with locals to extract private benefits, which have been shown to be actively promoted by local elites. For example, recent evidence from  show that due to the motivation to gain private benefits, elites in Nigeria create, evade and corrupt governance institutions. We, therefore, argue that foreign independent chairpersons (FIBCs) are an additional monitoring mechanism as their absence from the home country may help them overcome unethical CG practices compared to local chairpersons who are likely to be influenced by negative informal practices. Furthermore, existing literature suggests MNCs appoint foreign directors to increase governance isomorphism strategy through exportation of good governance guidelines and enforcement from host countries to EMs with weak governance enforcement. We contend, due to the international exposure of foreign chairpersons, they are more likely to understand and manage governance complexities and as such are better placed to promote EMMNCs' governance isomorphism objectives across economic environments.
Similarly, the extant literature (e.g., Oxelheim, Gregori, Randøy, & Thomsen, 2013) suggests directors from Anglo-American institutions improve CG disclosures quality. We propose that the recruitment of a foreign chairperson from countries with Anglo-American systems with robust CG enforcement (e.g., USA and UK) will enhance their ability to manage and improve boardroom efficiency. Therefore, they can reduce home country institutional weaknesses while simultaneously improve EMMNC's compliance with formal governance guidelines, which curbs the influence of antithetical governance practices.
More so, foreign chairpersons bring their human capital from their international experience to manage board processes and robust monitoring. Therefore, they can harness their socioeconomic diversity, varied institutional knowledge, professional, political, cultural and business backgrounds to enhance EMMNCs' governance practices. More so, host and home country experiences provides foreign chairpersons with 'first-hand' knowledge of global markets (Hahn & Lasfer, 2016;Masulis, Wang, & Xie, 2012) which they can harness to reduce firms liability of foreignness and improves legitimacy abroad while transferring good CG practices from other economic environments to enhance the quality of EMMNCs' CG practices at home. We contend that, the continual exportation of good governance practices by EMMNCs in the process of governance isomorphism through FIBCs improves the firm's competitiveness, visibility and success. Over time, this encourages peer firms in the home country to copy the CG practices of EMMNCs. In a repeated game of imitation of EMMNCs' CG practices (memetic isomorphism), peer firms in the home country develop resilient and more robust CG practices that are capable of bypassing negative informal practices (e.g., corruption) and weak regulatory enforcement of formal governance guidelines. Therefore, the evolution of EMMNCs' CG practices through governance isomorphism channelled by foreign chairpersons in addition to peer firm mimetic isomorphic practices may generate institutional change in governance practices in the home country (from left to right of Figure 1 through H1). We, therefore, hypothesize that H1: Ceteris paribus, there is a positive association between the presence of FIBCs and EMMNCs governance quality.

| CEO foreignness and the quality of CG practices
Like FIBCs, when foreign CEOs move to the home countries of EMMNCs, they carry along their human capital, resources, skills, knowledge and experience of other governance systems, which they employ in the day-to-day governance of the firm. As such, they can tap on their rich backgrounds and knowledge of different governance institutions to improve on the CG practices of EMMNCs at home in the process of governance isomorphism. Using board independence as an example, the Nigerian Securities and Exchange Commission (2011) governance code requires firms to have only one independent board member. Therefore, CEOs from South African or the UK with governance codes (King III report and UK 2010 and 2016 -Combine Codes respectively) that emphasis majority of the board to be independent are likely to transfer similar practices to Nigerian MNCs. This may lead to Nigerian MNCs meeting the Nigerian Securities and Exchange Commission (2011) governance requirement for board independence in addition to having more independent directors above the threshold required. More so, the extant literature suggests foreigners are less likely to engage in unethical practices such as bribery and corruption, which is prevalent in EMs Miletkov et al., 2017). Therefore, FCEOs are less likely to involve in corruption and elitism practices that can compromise their ability to adopt good CG practices and as such, can monitor the implementation of recommended CG practices than local CEOs. Furthermore, human capital literature (e.g., Clark et al., 2014;Fitzsimmons & Callan, 2016) suggest that CEOs bring their experience and knowledge from different firms and across different institutions to impact on the firm's governance practices. Therefore, FCEOs experience of governance institutions from their countries of origin and other host countries makes them more likely to reduce uncertainties and information asymmetry by diffusing good governance practices from abroad to improve on EMMNCs' governance quality. For example, compared to local CEO's, CEOs from the United States and the United Kingdom with more robust CG enforcement are likely to encourage transparency and accountability through the exportation of CG practices from their home country to improve on CG practices of Nigerian MNCs. More so, FCEOs are more likely to ensure the implementation of good CG practices that can overcome weak enforcement in the home country compared to local CEO's. Hence, FCEOs are more suited to promote EMMNCs adoption of formal governance guidelines rather than negative informal institutional practices (e.g., bribery and corruption).
We contend that continuous improvement of EMMNCs' CG practices through governance isomorphism by FCEOs in a repeated game curb unethical informal practice and generates a culture of good CG practices over time. The enhancement of EMMNCs' CG practices by FCEOs may improve their competitiveness and success at home and abroad in addition to building their reputation as successful examples in the home country. This may encourage peer firms to mimic and adopt their CG practices (from left to right of Figure 1 via H2). We therefore hypothesize that H2: Ceteris paribus, there is a positive association between the presence of FCEOs and EMMNCs' governance quality.

| Moderating role of board international interlocks
Resource dependency theory argues that boards of directors are an essential strategic resource for the firm in relation to its external resource needs. These linkages include networks with or affiliations to business elites, competitors, banks, as well as market and industry intelligence (Pfeffer, 1973). Therefore, firms appoint directors to their board to tap into the resources they bring from their external linkages. However, there is mixed evidence on whether board interlock with external environments improves on firm outcomes. For example, Falato, Kadyrzhanova, and Lel (2014) report that board interlock increases board busyness which is detrimental to monitoring quality and shareholder value. In contrast, Cai, Dhaliwal, Kim, and Pan (2014) report a positive effect of board interlock on information sharing and firm disclosure policies. Nevertheless, the impact of international board interlocks on firm practices is relatively unexplored in the prior literature. Due to mixed evidence and the lack of research thereof, we explore whether the international linkages of boardrooms affect the ability of agents of CG mobility to impact on EMMNCs' governance practices.
We argue that interlock with international boardrooms may affect the diffusion of CG practices from one country to another. Specifically, when directors seat in boardrooms in other countries, they bond with international CG practices which they can export to the home country to improve on EMMNCs' CG practices. More so, EMMNCs may appoint directors who seat on boardrooms out of the home country to benefit from their resource links in the host country as well as the director's ability to influence adoption of international CG practices. On the other hand, international interlocks can increase director's busyness and negatively affect their ability to attend board meetings and improve on firm CG practices. For example, the findings of Masulis et al. (2012) suggest that board busyness especially busy foreign directors impact negatively on board meetings attendance. This suggests that board interlock in foreign countries might be detrimental to board monitoring and, consequently, EMMNCs' governance quality.
We contend that for FIBCs to effectively monitor implementation and diffusion of good CG practices in the home country of EMMNCs, they need other board members especially those with international interlocks to support adoption of isomorphic governance practices. However, given that such directors may be busy with boardroom commitments in other countries, this may limit their ability to attend board meetings. Hence, the likelihood of FIBCs to affect the CG quality of EMMNCs diminishes. This effect may be more severe when there is a self-serving native CEO who may benefit from the director's absenteeism to influence unethical governance practices such as corruption and elitism. On the other hand, in cases where directors who seat on foreign boardrooms attend board meetings, they can assist FIBCs to improve and diffuse CG practices in the home country of EMMNCs. Therefore, board international interlocks may have either a negative or positive moderating effect on the ability of FIBCs to impact on EMMNCs' governance quality which reduces (improves) the likelihood of governance isomorphism (see H3 on Figure 1). We therefore hypothesis H3: Ceteris paribus, international board interlocks negatively (positively) moderate the association between FIBCs and EMMNCs' governance quality.
On the other hand, because FCEOs are present in the home country as they are involved in the daily governance of the firm, they are more likely to have more influence on EMMNCs' governance practices even in situations where directors are absent because of board commitment in other countries. Similarly, when directors who seat on foreign boards attend meetings, they collaborate with FCEOs to improve on EMMNCs' governance quality in the process of governance isomorphism. Therefore, cross-national interlock can positively influence the FCEOs ability to diffuse and improve governance quality of EMMNCs (shown in H4 on Figure 1). Hence, we hypothesis H4: Ceteris paribus, international board interlocks positively moderate the association between FCEOs and EMMNCs' governance quality.

| Context
Nigeria presents a rich institutional context for our study for several reasons. Recent World Bank statistics suggest that Nigeria is currently Africa's (and the World's 28th) largest economy with GDP (in 2017) of over $400 billion (Lange, Wodon, & Carey, 2018). It is also Africa's (and the World's 11th) largest oil producing nation. Nonetheless, the economy is reasonably well diversified. For example, despite its oil production pedigree, it only accounts for about 9% of the GDP as the country also boasts thriving services (55% of GDP), industrial (26% of GDP) and agricultural (18% of GDP) sectors (Lange et al., 2018). This makes Nigeria an attractive destination of foreign direct investment but also allows for the development of enterprises, which can compete on a global scale, that is, MNCs. Importantly, Nigeria bears the hallmarks of the typical EM, particularly in terms of high family control, concentrated ownership, weak national institutions and the prevalence of corruption and fraud (Adegbite, 2015).
To bolster its economic potential, international pressures from organisations such as the IMF and World Bank in the last few decades prompted the country to take on a programme of deregulation and economic liberalization (Ahunwan, 2002;Nakpodia, Adegbite, Amaeshi, & Owolabi, 2018). Proponents of these changes argue the potential for accelerating economic growth, development and alleviating poverty and corruption by enhancing responsible CG practices that are align to international standards (Adegbite, 2015;Ahunwan, 2002). As a result, the Nigerian Securities and Exchange Commission (SEC) has developed CG guidelines (including, SEC 2003 and2011 CG codes) providing recommendations for best practice of CG to be implemented by firms listed on the Nigeria Stock Exchange.
However, it is worth emphasizing that Nigeria is also a melting pot of over 500 different ethnic groups and consequently, diverse (and occasionally antagonistic) religious and cultural beliefs . Hence, the Nigeria SEC 2011 CG code emphasizes sensitivity to social and cultural diversity while also aligning with western CG systems. As discussed in , the United Kingdom and South Africa are preferred destinations for several Nigerian firms seeking cross-listing or cross-border expansion, perhaps, explaining the similarities between the CG codes of the three countries. It is in this context that this article explores the role of foreign leaders, specifically FCEOs and FIBCs, in shaping CG quality and practices within EMMNCs.

| Data sources and sample
Our sample covers multinational corporations listed on the Nigerian Stock Exchange (NSX) between 2011 and 2016. To identify multinational firms, we manually scan through financial reports of all listed firms to identify firms that report foreign operations. Due to the unavailability of the required CG data (especially country level CG disclosure data) for Nigerian firms in most databases (e.g., DataStream, Orbis, Boardex), we manually collect CG data from the annual reports of our sampled MNCs. We then collect financial (income statement and balance sheet) data for these firms from Thomson Reuters DataStream. We encounter problems with missing financial data, which we address by hand-collecting additional financial statement data. We then create our panel dataset by manually matching the data from DataStream to the hand-collected CG data using firm names and year. To be included in our final sample, firms must have complete annual reports covering the period of the study (2011)(2012)(2013)(2014)(2015)(2016). Our final sample consists of 80 unique multinational firms and a total of 480 firmyear observations. We focus on the 2011-2016 period for two reasons. Firstly, the SEC 2011 CG code became effective from the 2011 financial year, hence we measure CG quality using this guidelines for the post compliance period. Secondly, we limit our last year to 2016 because the Nigerian SEC introduced a draft revision of the 2011 CG guidelines in 2017. Therefore, to ensure consistency and avoid new and or future regulatory influences (e.g., changes in some CG provisions and applicability), we limit our last sample year to 2016. In other words, our choice of sample period is to ensure relevance and validity of measurement of CG quality to ensure consistency with SEC 2011 CG code.
We also included financial firms in our sample for two reasons. Firstly, these firms constitute more than 25% of listed firms in NSX and therefore represent a large section of the corporate sector in Nigeria. Secondly, compared to firms from other industries, financial firms in Nigeria have shown evidence of unethical practices in the past . For instance, top management teams of several Nigerian banks were imprisoned due to corruption and nepotism practices including allocating loans to their tribesmen, friends, relatives and themselves . To ensure that this choice does not bias our results, in our robustness checks, we exclude financial firms from our analyses. Additionally, in our analyses, we control for industry effects. As we will discuss later, all our results are robust to this choice.

| Dependent variable
Our dependent variable is corporate governance quality (CGQ) which is measured as a firm's total compliance to the 75 governance provisions required by the Nigerian Securities and Exchange Commission (2011) CG code. In contrast with CG codes in other countries with some provisions applicable to premium or large listed firms only (e.g., 2016 and 2018 UK Corporate Governance codes), all the provisions of the Nigeria SEC 2011 CG code are applicable to all firms listed on the NSX irrespective of size or industry. In line with prior research (e.g., Gyapong & Afrifa, 2019; Ntim, Opong, & Danbolt, 2012), each provision is measured as a dichotomous score of one or zero. The final CGQ score is a continuous variable that ranges from 0% up to 100% (full compliance with all the provisions of the SEC code). For example, a firm that adopts 51 of the 75 provisions in a year has a score of 68%.

| Explanatory and moderating variables
Our first explanatory variable is FIBCs measured as a dichotomous variable, which takes the value of one if the board chairman is a non-executive director and nonnative of the firm's home country (non-Nigerian), and a value of zero, otherwise. Similarly, we measure FCEOs as a binary variable with the value of one when the CEO is not a native of the home country (Nigeria), and a value of zero, otherwise. Our moderating variable, board international interlock (BII), is measured following prior research (e.g Ruigrok, Peck, & Keller, 2006;Stuart & Yim, 2010) as the number of board seats occupied by directors out of the firm's home country.

| Control variables
Prior literature (e.g., Hooghiemstra, 2012;Ruigrok et al., 2006) suggest that several CG variables impact on governance quality. Hence, our analyses controls for such influences. First, research (e.g., Boulouta, 2013;Cai et al., 2014) suggests that, due to the ethical behavior and diversity of ideas brought into boardrooms by female directors, they improve on board decision-making and firm CG practices. Hence, we control for boardroom gender diversity using the percentage of women on the board (GD). Independent boards are more likely to scrutinize firm compliance with CG regulations than insider dominated boards (Cornett, McNutt, & Tehranian, 2009;Khan, Muttakin, & Siddiqui, 2013). This suggest that boards with high proportion of non-executive directors (NED) are likely to have enhanced CG quality. We control for board independence using the percentage of NEDs to the total board size and proportion of independent directors in the audit committee (ACI). Furthermore, the separation of CEO and Chairman position (CEO duality) has been argued as necessary for effectiveness of CG practices (Fitzsimmons & Callan, 2016;Tian et al., 2011). Specifically, when the position of CEO and chairman is separated, there is enhanced monitoring and control which leads to improvement in CG quality. Hence, we control for CEO duality (DL) using a dummy variable which takes a value of one if there is separation of CEO and chairman positions and a value of zero, otherwise.
The presence of institutional shareholders ensures stronger incentive to monitor CG practices than do individual investors as they have larger stakes in the firm (Chung & Zhang, 2011). This suggests that the presence of institutional investors reduces information asymmetry and improves CG quality. We control for institutional shareholding (ISH), defined as the proportion of shares held by banks, mutual funds and insurance company's to the total value of shares. More so, firm performance has been shown to affect firm's ability to adopt recommended CG practices as compliance to guidelines require significant expenditure (Gaur et al., 2014;Westphal, Seidel, & Stewart, 2001). We thus control for firm performance using return on asset (ROA) and Tobin Q (Q). Furthermore, large and fast growing firms are more likely to have adequate resources to enable compliance with CG regulations than their smaller and low-growth counterparts (Ntim, Lindop, & Thomas, 2013). Hence, we control for size and growth influences using capital expenditure as a percentage of the total asset (CAP) and sales growth (SG). Lastly, we also control for industry and year effects using six industry and year dummies respectively. Definitions and measurements of variables are presented in

| Estimation methods
Prior evidence (e.g Roberts & Whited, 2013;Schultz, Tan, & Walsh, 2010;Wintoki, Linck, & Netter, 2012) suggests that endogeneity is a serious issue that can affect inferences from research of this nature. To ensure our results are not biased to endogeneity, we employ a three-stage least square (3SLS) estimation method that has been evidenced to control for endogeneity (e.g Denis & Sibilkov, 2010;Estélyi & Nisar, 2016). Before conducting a 3SLS estimation, we use the Durbin-Wu-Hausman homogeneity test (see Larcker & Rusticus, 2010, for detailed discussions on how to implement instrumental variable models) to examine whether there is a simultaneous link between our dependent and independent variables. The findings reject the null of no endogeneity, suggesting that foreign leadership and MNC CG quality are endogenously related.
This suggests that OLS may produce bias estimates and therefore 3SLS is a more suitable estimation method. However, for robustness, we also present results from the standard OLS estimation. Our 3SLS estimation equations are stated as; In Equation 1, CGQ is predicted by foreign independent board chairmanship (FIBC), foreign CEO (FCEO) in Dichotomous variable which takes the value of one if the board is chaired by a non-executive director who is a non-Nigerian and a value of zero, otherwise.
Foreign chief executive officer (FCEO) Dichotomous variable which takes the value of one if the CEO is a non-Nigerian director and a value of zero, otherwise.

Board international interlock (BII)
Number of boards seats occupied by directors out of the firm's home country.

Panel C: Control variables
Gender diversity (GD) Female directors as a percentage of number of board members.

CEO duality (DL)
A dummy variable one, if there is separation of CEO and chairman positions and a value of zero, otherwise.
Non-executive directors (NED) Non-executive directors (NED) as a percentage of board size.
Institutional shareholding (ISH) Percentage of institutional shareholding to the total shares of a firm.

Audit committee independence (ACI)
Non-executive directors as a percentage of number of audit committee members.
Return on asset (ROA) Earnings for the year as a percentage of total asset.
Tobin q (Q) Ratio of total assets minus equity book value plus market value of equity to total assets.
Capital expenditure (CAP) Capital expenditure as a percentage of firm total assets.

Sales growth (SG)
The difference between current and previous year sales revenue as a percentage of previous year sales revenue.

Industry dummies (IND)
Dummy for each industry group (total of six industry dummies).
Year dummies (YD) Dummy for each of the six firm years.  addition to the moderating variable (BII) and 11 control variables, including gender diversity, audit committee independence, institutional shareholding, Return on Asset (ROA), Tobin Q (Q), sales growth (SG), capital expenditure (CAP), CEO duality (DL), proportion of NEDs (NED), industry dummies (IND) and year dummies (YD). Equation (2) re-examines Equation (1) but with the addition of interaction effects between FIBCs and international interlock (FBC*BII) and the latter with FCEOs (FCEO*BII).

Variables
To conduct 3SLS, we need instruments that meet both the sufficiency and validity conditions. Institutional theory suggests that, the institutional environment influences the behaviour of economic actors (Cumming et al., 2017;Dauth et al., 2017). This implies the behaviours of foreign chairpersons and CEOs are influenced by the institutional peculiarities of corporate practices in their countries of origin including corporate ethics and accountability. Corporate ethics and the level of accountability of FCEOs and FIBCs countries of origin is collected from the World Economic Forum Global Competitive Index datasets. In untabulated correlations, we find that corporate ethics and accountability in the country of origin are highly correlated with FIBCs (coefficients of 0.96 and 0.98 respectively). Similarly, the country of origin corporate ethics and accountability are strongly correlated with FCEOs (coefficients of 0.97 and 0.98 respectively). These suggest the instruments meet the sufficiency condition. Furthermore, these instruments are uncorrelation with the error terms on the second stage, which implies they meet the validity condition. More so, we also conducted the Hanson-Sargan test of over-identification, and the p-values are above 0.32 suggesting the instruments meet the exclusion condition, which also confirms the instruments are exogenous.

| Descriptive statistics and correlation diagnostics
The descriptive and correlation statistics are reported in Table 3. The average CGQ of the sampled MNCs is 73.70% suggesting that on aggregate, firms are adopting recommended CG practices. This is comparable but moderately higher to CG quality of 71.33% reported by Ntim et al. (2013) for King II CG compliance level but significantly higher than the 61.44% compliance to King III guidelines recently reported by Gyapong and Afrifa (2019) for South African firms. However, while some firms comply with up to 100% of the governance guidelines, others comply with about a quarter of the guidelines. Also, there is high variability between firm-level CG quality (17.12%).
Foreign CEOs manage approximately 28% of the sampled MNCs. Regarding board leadership, about 11% of boards are led by foreign chairpersons. More so, on average, the sampled MNCs have at least one interlock with boards in a foreign country. These results compares favourably against results from other contexts. Specifically, our sample firms report significantly higher levels of average director foreignness compared to firms in China, Kenya, India, Mexico, Brazil South Africa and Zimbabwe (Estélyi & Nisar, 2016). Indeed, besides the USA, our results when assessed against the 30 countries (both developed and developing nations) examined in Estélyi and Nisar (2016), suggest that Nigerian MNCs are among the highest in promoting appointment of foreign leaders to manage their operations and boardrooms.
The control variables also show high variability. For example, the average of NEDs is 71.63% with variability of 12.66%. This is significantly higher than that reported for Nigeria (21%) by Ehikioya (2009) for the period 1998-2002. Similarly, 98% of the firms have separated the CEO and chairperson positions with a variability of 13.5%. This also indicates that separation of leadership position has improved from 91% between the 1998 and 2002 period (see details in Ehikioya, 2009, p. 236). Overall, these descriptive statistics suggest that the various Nigeria SEC CG codes have improved CG standards in listed Nigeria firms over time. Finally, on average, ROA is 3.6 with a standard deviation of 12.95.
Correlation results are presented in Columns 4-16 on Table 3. The pairwise correlations are generally low to moderate (with a maximum of 0.44), suggesting that multicollinearity is not a major concern in our subsequent estimations. For additional robustness, we examined the variance inflator factor for each of our equations and the highest is 2.59, which is below the critical value of 10. We also examine Cook disturbance and tolerance statistics (not reported for brevity reasons), and they all assuage multicollinearity concerns. Interesting, CG quality is significantly associated with both FIBCs and FCEOs, which provides early indications in support of our main hypotheses. Table 4 presents results from our hypotheses tests. In Model 1, we examine H1 and H2. In Model 2, we add the moderating effect of international interlocks to assess H3 and H4. Our main estimation method results (3SLS) are reported in Columns 2 and 3, and OLS results are reported in Columns 4 and 5.

| Main results and discussion
First, we hypothesis (H1) that, FIBCs will impact on the governance practices of EMMNCs. As shown in Table 4 for both 3SLS and OLS, the results are significant and consistent with H1 suggesting that FIBCs positively impact on EMMNCs' governance quality. In terms of economic significance, the appointment of a foreign board chairman is associated with approximately 5.8% improvement in the CG quality of EMMNCs. Thus, the results support our argument that in the process of governance isomorphism, FIBCs utilize their knowledge of varied governance institutions including different cultural and business backgrounds to enhance the CG practices of EMMNCs which may evolve overtime to resilient governance practices and a source of mimicking for other firms. The finding is consistent with recent advances in CG research by Estélyi and Nisar (2016);Miletkov et al. (2017) who also evidenced that foreign directors improve on the CG practices and corporate outcomes of firms especially in countries with weak governance enforcement. Similar to FIBCs, in the second hypothesis (H2), we propose that foreign CEOs (FCEOs) will positively influence the CG quality of EMMNCs. The results from Table 4 supports this conjecture that FCEOs enhance MNCs governance quality at home, which may evolve to more resilient CG practices and become a source of imitation by peer firms. Specifically, the appointment of FCEOs improves CG quality of EMMNCs by approximately 7.58%. This result is consistent with the recent findings of Chuah and Foong (2019) that FCEOs improve corporate outcomes but contrast the result of Vo, Nguyen, Tuan, Luu, & Vu (2020) that foreign CEOs under-perform in countries with good governance supporting institutions. Nonetheless, our finding indicates that non-native CEOs improve the CG quality of EMMNCs due to their experience and knowledge from different governance institutions. We contend that, the experiences and knowledge of governance institutions from their home country and other host countries increases the likelihood of transparency and diffusing good governance practices internationally to improve on MNCs governance quality in weak regulatory environments.
Furthermore, in the third hypothesis (H3), we suggest that interlocks with foreign boards moderate the impact of FIBCs on EMMNCs governance quality. Consistent with our prediction of H3, the coefficient of the interaction between foreign board chairs and international interlocks (FIBC*BII) is significantly and negative in both 3SLS (Column 3, Table 4 and OLS Column 5, Table 4). In economic terms, the appointment of foreign board chairperson in a highly internationally interlocked boardroom reduces the impact of FIBCs on CG quality of EMMNCs by approximate 4.33%. This implies board interlocks out of the home country of EMMNCs negatively affects the ability of FIBCs to impact on the firm's CG quality at home. This result is consistent with the busyness argument of interlocking directors and in support of Masulis et al. (2012) results that other board commitment out of the company may limit the ability of directors to attend board meetings which reduces their ability to assist the board chairman in enhancing the CG quality of firms.
Finally, in the fourth (H4) hypothesis, we argue that due to the presence of foreign CEOs in the daily governance of firms, they are less likely to be affected by the busyness of directors due to interlocks with foreign boards. Hence, we proposed a positive impact of foreign interlocks on the association between FCEO and EMMNCs' governance quality. This prediction (see Columns 3 and 5 of Table 4 for 3SLS and OLS estimations respectively) is supported. Specifically, the appointment of a foreign CEO in boardrooms with foreign interlocks improve the CG quality of EMMNCs by approximately 2.27%. This suggests international interlocks enhance the likelihood of FCEOs to strengthen governance practices of EMMNCs through the repeated game of governance isomorphism. Our results is consistent to the findings of Chuah and Foong (2019);Field, Lowry, and Mkrtchyan (2013) and suggest that, despite the busyness of interlocked directors, their advisory due to the experience in foreign boards enhances the ability of foreign CEOs in weak governance enforcement environments to improve on the CG practices of firms.

| Robustness test
So far, we have shown that our results are robust to estimation methods. Also, our 3SLS estimation evidence that our findings are robust to endogeneity from simultaneity and unobserved firm-specific factors. The 3SLS model is plausibly more robust than its two-stage least squares (2SLS) alternative, as it has additionally controls for cross-correlations of residuals. Nonetheless, for confirmatory reasons, we re-examined our hypothesis using the 2SLS regression specification. The results are shown in Columns 2 and 3 of Table 5. As can be seen, the results confirm our reported findings.
Furthermore, some scholars (e.g., Pham, Suchard, & Zein, 2011;Roberts & Whited, 2013) argue that 3SLS does not control for dynamic endogeneity. For example, past values of the dependent variable can affect both current values of the dependent and independent variable. For instance, poor CG quality in the previous year can coerce EMMNCs to appoint FIBCs and FCEOs to improve on future CG quality. As a result, we conduct dynamic system GMM estimation which has been shown (see for example Wintoki et al., 2012) to control for dynamic endogeneity. The results are presented in Columns 4 and 5 of Also, CG scholarship argue (see e.g., Ntim et al., 2012) that directors are more inclined to adopt governance guidelines that have a direct impact on firm returns than those that address stakeholder needs but has no direct impact on firm performance. To ensure our results are not sensitive to aggregation of these provisions in our CG quality index, we split the index into two subindices with 61 provisions that address accountability to shareholders (SCGQ) and 14 provisions (SKCGQ) that report on stakeholder governance. The shareholder (SCGQ) and stakeholder (SKCGQ) sub-indices are dependent variables in Columns 3-4 and 5-6 on Table 6 respectively. As can be seen, our results remain robust across both sub-indices.
Finally, as financial firms contribute towards a large proportion of our sample, we examine whether our results are robust to the exclusion of these firms. These results are presented in Columns 7 and 8 of Table 6. Still, our results remain qualitatively similar and our conclusions unchanged. This suggests that our results and conclusions are robustness to the inclusion of financial firms.

| Summary of findings
Motivated by the dearth of research that explores whether foreign leadership of emerging market multinationals impact on the quality of their CG practices, we examine the effect of foreign chairpersons (FIBC) and CEOs (FCEOs) on the CG quality of EMMNCs. Also, we assess whether these associations are moderated by interlock of EMMNCs boardrooms with foreign boardrooms. Using Nigerian MNCs as exemplification of EMMNCs, our findings are consistent with our main argument that foreign leadership improve on the CG quality of firms in weak governance enforcement environments. Specifically, FIBCs and FCEOs positively impact on the CG quality in the home country of EMMNCs. Furthermore, we also find that the effect of foreign chairpersons on the governance quality of EMMNCs is negatively moderated by international interlock of boardrooms. On the other hand, interlocks with boardrooms out of the home country of EMMNCs positively and significantly moderate the impact of non-native CEOs on EMMNCs' governance quality. Drawing on these findings, we contribute to CG literature across several dimensions.
First, we develop a conceptual framework (Figure 1) that extend institutional isomorphism and repeated game theories by uncovering how EMMNCs can be a source of institutional evolution of CG practices in EMs through foreignness of their leadership. Specifically, we show that the motivation to reduce external dependencies in host countries because of internationalization coerces EMMNCs to adopt isomorphism strategies that ensure legitimacy, reduce the liability of foreignness abroad in addition to reducing institutional weaknesses of the home country. The implementation of isomorphism strategies by EMMNCs is achieved through the appointment of foreigners to lead the firm and the board. These foreign leaders enhance CG quality in the home country of EMMNCs through governance diffusion. We contend that the repeated game of importation of good CG practices to EMMNCs through their leadership evolves to more resilient governance institutions and becomes a source of mimicking by local firms. The repeated game of imitation of EMMNCs practices by peer firms in the home country evolves to institutional change in CG practices in EMs that is capable of bypassing both weak enforcement of formal governance guidelines and unethical informal practices (e.g., corruption) which is prevalent in EMs.
Second, we advance the growing literature on CG mobility. We note that studies in this area have mostly  (1) can be correlated but no serial correlation should exist in the second difference AR (2). Hanson test of over-identification (J-Statistic) is a chi-squared distribution with (l − r) degrees of freedom with l indicating the number of moment conditions and r, the parameters to be estimated; with a null (H0) = moment conditions are specified correctly (i.e., instruments in the dynamic system GMM are valid). Diffin-Hansen tests of exogeneity has a null (H0) = instruments in the system equations are exogenous. Full variable definitions are available in Table 2. concentrated on proportion of foreign directors (e.g., Estélyi & Nisar, 2016;Miletkov et al., 2017) or foreign shareholders (e.g., Aggarwal et al., 2011) and bonding of firms through dual listing (e.g., Temouri et al., 2016). We contribute to advance these studies, but in contrast, we explore and provide novel evidence that Note: The table presents the impact of chairperson and CEO foreignness on CGQ sub-indices and and the moderating role of international interlocks and excluding financial firms. ***, **, * denotes significance at 1, 5 and 10% levels respectively. Robust mean square standard errors (RMSE) are in parentheses. SCGQ and SKCGQ represent sub-indices measuring compliance with 61 shareholder-oriented and 14 stakeholder-oriented CG recommendations of the 2011 Nigeria SEC code of good practices in CG. Full variable definitions are available in Table 2. foreign leadership of EMMNCs serve as channels of governance mobility. Specifically, we draw on human capital and board leadership independence perspectives to uncover how the international dimension of boards and firm leadership leads to CG mobility across economic environments. Specifically, we show that FCEOs and FIBCs are important channels of governance mobility that positively impacts on EMMNCs' governance quality at home. Third, we extend board interlock literature (Cai et al., 2014;Chuah & Foong, 2019;Falato et al., 2014;Field et al., 2013) by incorporating an international perspective of the latter and showing how this affects the ability of agents of CG mobility to impact on MNCs governance quality at home. Specifically, we evidence that, while EMMNCs interlock with foreign boards may advance the ability of FCEOs to export international CG practices to EMs, it negatively affects the likelihood of FIBCs to diffuse CG practices due to directors busyness with boardroom commitments in other countries.
The findings of this study also has several managerial relevance. Firstly, for EMMNCs, we offer them incentives to appoint foreigners to lead the firm and boards. We show that foreign individuals who occupy chairmanship and CEO positions do not only improve legitimacy and reduce the liability of foreignness in the host country but also promote governance diffusion practices which enhances the quality of CG practices as recommended by regulators in the home country. Also, we believe that improvement in CG practices may assist EMMNCs to compete with international rivals and may improve on their visibility as success stories in the home country. This can reduce the uncertainties and information asymmetry concerns that foreign investors may have when investing in EMs. Therefore, this may attract an inflow of capital from foreign shareholders who are more likely to invest in firms with quality CG practices. Finally, we highlight the importance of EMMNCs and foreign leadership as mechanisms of institutional evolution and change in EMs. Specifically, when EMMNCs employ foreign leaders to manage their operations and board functioning, they bond with foreign CG practices that can overcome weak enforcement and negative informal practices (e.g., corruption) which may evolve to resilient governance institutions and bring about institutional change in EMs through mimicking by local firms.

| Limitations and future research implications
Our research may suffer from certain caveats that should be examined by future research. First, while the theoretical and practical insights of this study may apply to other EMs and assist in unravelling how EMMNCs can bring institutional evolution and change in their countries of origin, it may be interesting to explore the extent to which the findings are generalizable across contexts. Specifically, because we examine MNCs with origin from a single emerging economy, this may affect cross-country generalization of results. Furthermore, our research provides insights on foreign leadership as channels through which EMMNCs can improve their governance practices at home. Nevertheless, these governance mobility agents and those reported so far in existing literature may not be the only international governance mechanisms that firms in EMs can use to bring institutional change. For example, native directors and managers in EMs who have studied or worked in foreign countries (especially in countries with robust corporate systems) can serve as bonding mechanisms to international CG practices. Hence, returnee directors may be additional channels of institutional change as they can use their knowledge from foreign countries to improve on EMMNCs and non-MNC CG practices. Hence, we suggest future research to examine other channels of governance mobility and institutional evolution in EMs beyond those discussed in the literature.