ESG Reporting: Empirical Analysis of the Influence of Board Heterogeneity from an Emerging Market

Firms are facing pressure to convincingly communicate to stakeholders their environment, society, and corporate governance (ESG) disclosure. In developing countries, where frictions among controlling and non-controlling shareholders are pervasive, the possible dissensus inside boards regarding ESG disclosure remains understudied. We investigate the ways in which boards’ heterogeneity between the interests of controlling groups and the interests of institutional investors influences ESG disclosure of firms in the Latin American context. Using social networks and logit panel data models, we analyze for 2015-17 the probability of ESG disclosure by 124 Chilean listed firms. Our evidence suggests that the influence of controlling shareholders through directorate interlocking has a negative relation with ESG disclosure. Additionally, we observe that the influence of institutional investors on ESG disclosure is not yet critical. Moreover, we find partial evidence of the presence of tension within the boards regarding ESG reporting between the directors that represent controlling shareholders and institutional investors. Considering the importance of institutional investors and the ubiquity directorate interlocking among Latin American’ firms, our results are relevant for regulators involved in advancing the rules of ESG disclosure practices, institutional investors focused on enhancing their ESG investment strategies, and firms engaged in improving the ESG decision-making within their boards.


Introduction
The belief that conducting business in a sustainable way does not exclude fostering a suitable environment entices good governance of firms, and promoting good social practices seems not to be a short-lived fad; instead, evidence indicates that it appears to be a long-term trend. Nowadays, concern for the impact of business decisions on the environment, society, and corporate governance quality (ESG criteria) is not only a crucial element of the agenda of governments and regulatory institutions; it is also a critical component of the decision-making of the most important asset managers around the world [1].
Worldwide, the forces behind the attention that ESG performance receives are multiple and respond to many elements. For instance, the societal change that involves the transfer of wealth from baby boomers to new generations of investors who are more sensitive to ESG aspects causes the preferences of these new investors to affect the asset allocation criteria of institutional investors. Likewise, now at a market level, there are factors linked to supervisory issues that also contribute to this trend. Such was the case after the subprime crisis: it led to the emergence of greater regulation of investments, new managerial practices, and novel rules set by governments and stock exchanges [2].
In this context, the regulation of the public dissemination of ESG performance information through mandatory firm reports in conjunction with the disclosure of ESG managerial the resolution of which produces the firm's ESGD decision. Moreover, we argue that in our emerging market setting, this tension occurs because the position (regarding ESGD) between controlling shareholders and institutional investors shareholders differs, which we specifically test. The focus and conjectures of our study are justified by the fact that in emerging markets, and particularly in Latin America, there are key characteristics that provoke this tension through concentration of property. These key characteristics are a weak legal enforcement of minority shareholders compared to developed countries, less financial development than in developed countries, and limited liquidity in capital markets.
As the literature states, shareholders do not exclusively receive cash flows in exchange for their money, but they do receive control rights [10]. For instance, they have the right to vote at shareholders' meetings, to demand that a court enforces their rights, and to claim ownership of company assets left as collateral [11]. Thus, in the presence of friction, such as agency problems or information asymmetries, or weak property rights, these shareholder rights become fundamental to the firms in order to obtain external financing, as they protect minority shareholders from any expropriation or opportunistic behavior from the controlling groups of the firm. The foregoing explains the likelihood of conflicts of interest between majority and minority shareholders in both developed and developing markets. These conflicts of interest are more prevalent in less developed financial markets [12].
In addition, frictions within the board could also be the result of the property structure of a firm. The weaker legal environment that firms' owners face means that, in developing countries, there is a greater concentration of the property of firms in a large economic group than in developed markets [13]. An implication of this is the ample evidence of directorate interlocking (DI) among firms. As the literature states, in emerging markets, DI is a strategy used by controlling shareholders as a way to improve their control over firms as well as to enhance firms' profitability and performance. For instance, facing scarcity of financing, firms opt for using DI for lowering their costs of access to credit [14][15][16].
The two previously mentioned characteristics (firms controlled by economic groups and the presence of interlocking between company boards) are present in emerging markets, and particularly in the Chilean market. Moreover, in the Chilean financial market, there is a growing presence of institutional investors acting as non-controlling shareholders of the listed firms on the Santiago Stock Exchange. Over the last 30 years, these professional investors have coexisted with the controlling economic groups of the firms, balancing their long-term investor role with the predominant profit-maximizing ethos of these economic groups [17]. However, the control of Chilean publicly traded firms is not undisputed because of the participation of institutional investors; under the Chilean law, they act as minor shareholders.
One of the main contributions of our paper is to expand the study of eventual tensions between groups of board members and ESG disclosure. In Latin America, economic groups act as controlling shareholders and have a relevant influence in the boards' community. This occurs because they appoint directors of their confidence in several firms simultaneously. For this reason, we use social network measures of the directorate interlocking network to capture this margin of influence of controlling shareholders, i.e., economic groups. On the other hand, the participation of minority investors such as institutional investors has increased during the last decade. We argue that this heterogeneity is relevant because it may cause a tension between the interests of the controlling shareholders and the interests of the minority shareholders to disclose ESG performance. The lack of empirical evidence on ESG disclosure in Latin America, coupled with the peculiarities of the business environment, prevents a direct extrapolation of the evidence obtained from developed countries [5].
The particularities of the Chilean market transform it into a unique and vibrant environment to study whether there is a tension regarding the ESGD decision between noncontrolling shareholders through institutional investor participation and the controlling shareholder via interlocking directorate strategy. First, in Chile, it is possible to combine the unique characteristics of the corporate governance of developing countries with the presence of relevant institutional investors in the region. Second, as a recent OECD (Or-ganisation for Economic Co-operation and Development) member, the country possesses high-quality public information and has extensive availability of financial data of firms. In this way, we hope that the results of our research will be useful for investors who wish to expand their knowledge of the ESG phenomenon in emerging markets; for market regulators who wish to refine regulations for the reporting of companies' ESG activities; and for the firms themselves, which must make the decision to incorporate directors into their boards in order to lead their ESG initiatives.
The remainder of this paper is organized as follows. Section 2 reviews prior research and the development of the hypotheses. The methodology, data, sample, and empirical strategy are described in Section 3. Section 4 presents the analyses, results, and discussion. Finally, Section 5 presents our conclusions.

Theory and Hypotheses
After the great financial crisis of 2008-9, ESG performance and reporting emerged as relevant issues in financial markets, the business world, and the academy. According to the OECD [1], the way in which institutional investors address ESG issues is receiving increasing attention across its member countries. Worldwide, professional investors (such as mutual funds, insurance companies, and pension funds, among others) are under increasing pressure to understand, manage, and respond to potential hazards and opportunities arising from ESG factors. In this sense, ESGD is a critical input that enables them to incorporate these elements into the decision-making process of an institutional investor. Consequently, investors will seek to either avoid or reduce exposure to those assets that pose greater ESG risks or to influence firms to make them more ESG-friendly and thus generate more positive benefits for society [2]. Currently, ESG investing represents a growing portion of overall capital market investments, reaching USD 30 trillion in sustainable assets under management (AUM) worldwide by the end of 2019 [18], and institutional investors that signed the Principles for Responsible Investment (PRI) ran more than USD 80 trillion of AUM at the end of 2019 [2].
The emergence of ESG practices also impacted corporate management behavior. For instance, if a company fails to treat its customers or employees properly, or if it ignores the impact of its operations on society or the environment, future cash flows, and thereby equity value, are likely to be negatively impacted. As a consequence, the firms and their boards have to balance the trade-off between short-term maximization profits versus longterm value maximization [18]. Recently, this tension emerged in the media, evidencing the discussion about shareholders' value maximization vs. stakeholders' value maximization (a statement signed by the Business Roundtable [19] in August 2019 by numerous leading CEOs stated the importance of all stakeholders: "Each of our stakeholders is essential. We commit to deliver value to all of them, for the future success of our companies, our communities, and our country". In the same vein, in 2020, the CEO of BlackRock-one of the largest asset management firms in the world-stated that, "A company cannot achieve long-term profits without embracing purpose and considering the needs of a broad range of stakeholders"). These opinions from leaders in the corporate and financial worlds seem to indicate that, nowadays, the traditional maximization firm's objective needs to be adapted to also include ESG objectives.
From the literature's perspective, there are at least three reasons to study what moves firms to engage in ESGD practices. First, ESGD is a way of showing that firm operations are within societal limits, and hence that firms' activities are legitimate [9]. Thus, ESGD serves as an input for managers that helps them to enhance ESG performance, evaluate the impact of their decisions on stakeholders, and serve as groundwork for preventing negative shocks [1, 17,18]. Second, ESGD focuses on the need for firms to reduce the asymmetries of information between their controlling and non-controlling shareholders, in addition to helping policymakers and governmental entities in the task of promoting transparency and monitoring the impact of companies on society [20,21]. Finally, ESGD relates to the multidimensional elements behind firms' decisions and how the academy can learn about management s decision-making [9,22].

Legal System and Corporate Governance in Emerging Markets
The existing tensions in the corporate governance of firms inherent in the relationships within the board of directors are relevant factors to explain the value of a company and its performance over time. From the financial perspective, in a frictionless world, the value of a firm would depend exclusively on the future cash flows that it generates and its associated risks. Hence, in this world, property rights that, for instance, protect shareholders from opportunistic behavior from the controlling groups of the firm are irrelevant. Nonetheless, in the real world, they are relevant factors for explaining the value of firms.
The effectiveness of property rights for investors is based on two factors. The first corresponds to the existence of rules and laws that provide adequate protection, and the second to the degree of application of these regulations and laws in the markets and countries where capital assets are issued. La Porta et al. [23] indicate that the differences in the quality of laws and their applicability-legal enforcement-explain the ability of firms to raise capital and the willingness of investors to finance them. They point out that a direct consequence of the above would be differences between countries and markets in the capital structure of companies and the concentration of their ownership. Thus, the legal protection of investors and the applicability of the rules would play a fundamental role in how companies are financed and in how shareholders participate in property in the capital markets [13].
Although the laws for legal protection for investors and the degree of applicability of the law between countries may differ, evidence shows that their legal origin does not [13]. There is ample evidence of differences in the legal protection of investors between countries based on English law (Common Law) compared to those based on Roman (Civil Law). La Porta et al. [13] show that investors are least protected in Civil Law countries, and especially in those of Roman legal origin, while Common Law countries provide the greatest protection. In addition, they indicate that the quality of enforcement in legal systems based on Common Law is greater than in legal systems based on Civil Law. The relevance of the above is that lower investor protection results in a greater concentration of company ownership [13], lower stock valuations [24], less access to external financing [25], lower equity market liquidity [26], and less financial development [27].
A low-legal-protection environment forces investors to seek mechanisms that reduce these shortcomings in the legal environment. La Porta et al. [23] indicate that there are three ways to deal with weaker legal protection. The first is promoting the higher quality and applicability of existing laws; the second is introducing modifications to the current regulations in order to minimize the probability of expropriation from investors, and the third is increasing the concentration of ownership of the firms in order to have stronger and more organized shareholders who can monitor the majority shareholders and the top management of the companies.
The financial literature shows that improvements in corporate governance, that address the consequences of lower legal protection for shareholders, are beneficial to the firms. In the first case, the evidence shows that, in emerging countries, increasing access to financing reduces the cost of capital, improves performance, and favors better treatment for all stakeholders. In the second case, it produces a better allocation of resources in the economy, thus supporting one of the primary functions of markets, with the consequent positive effect on the financial development and growth of countries [28][29][30][31][32].
The above results are especially relevant in the emerging market context. Despite the paucity of literature, evidence indicates that compared to developed markets, emerging ones have limited financial development, less access to external financing, highly concentrated ownership structures of firms, and low participation of institutional investors in the ownership of the listed firms [29]. The Latin American environment is no different from the rest of the developing countries. Le Fort [33] points out that corporate governments in the region must deal with the high concentration of ownership of firms and the existence of large economic conglomerates. This would affect the corporate structure of the firms and generate an agency problem between controllers and the rest of the shareholders.

Directorate Interlocking
For over four decades, different studies have suggested that social relations of individuals within and between organizations affect various dimensions of organizational performance and decision-making. Corporate linkages between firms caused by sharing a common director or board interlock [34] have been the subject of study in various disciplines. For instance, sociological, social network, and economic analyses have been widely used to study the origins of directorate interlocking (DI) [35][36][37][38][39]. Other studies have focused on various outcomes of DI, such as reducing the uncertainty of the firm [40], accessing information [41], diffusing strategy and practices [42], and influencing the perception of firm strategy and quality [43,44].
Recently, based on a broad review of the literature, Lamb and Roundy [45] indicated the main aims for DI from the perspective of firms and boards (in terms of the consequences of DI, they pointed out five classes of outcomes studied in the literature, namely reduction of environmental uncertainty and its effects [39,46,47], gaining access to diverse and unique information [30,40,48], disseminating strategies and practices [49,50], shaping perceptions of the firm's quality [42,43], and influencing firm performance [51][52][53][54]). DI helps firms to obtain resources, facilitates monitoring activities on subsidiaries, can send signals to potential investors, or may gain access to the specific human capital of a particular director; additionally, DI helps boards to monitor a firm's executives and provide resources to the firm.
Related to boards' duties, Hillman and Dalziel [55] point out two main functions: monitoring the management team on behalf of shareholders and providing resources. By minimizing agency costs, the monitoring duty improves the performance of the firms. On the other hand, as resource dependence theory states, the board acting as a resource provider enhances performance too. These resources, known as board capital, are delivered in the form of advice and counsel, legitimacy, links to other organizations, and connections to the establishment and mainly consist of both the human and relational capital of the directorates. Specifically, the resources provided by the board are based on networks to strategic firms and institutions and to the expertise, reputation, and experience of the board members.
Although DI aims to create value for the firm, we argue that in the legal context of Latin America, the concentration of property causes DI to respond to the governance strategy of economic groups. Thus, the intensity of DI captures the influence of controlling shareholders at the firm level. Accordingly, the contemporaneous correlation between DI and the adoption of ESG management and reporting practices may vary depending on the preferences of the controlling shareholders in the market. That is, if the preferences of the controlling groups coincide (do not coincide) with ESG models for companies' behavior then we should observe a positive (negative) relationship between the DI level of firms and their ESGD.
The latter is especially relevant in emerging market contexts due to a scarcity of financing. The literature states that in developing markets, there is a shortage of credit to firms, which is reflected in short-term economic and financial performance, with this phenomenon being one of the main causes of DI [14] (evidence indicates that firms facing credit constraints are in a better position when their boards have relevant connections with funds providers. Braun et al. [14] evidence an important role of DI as a device for dealing with credit constraints and providing access to financing. They show that firms connected to banks exhibit better access to credit-higher leverage ratios, greater valuations-higher market-to-book ratios, and better chances of surviving as measured by a greater survival time). There is also evidence that firms that are well connected at the political level have better access to financing. Recent papers have documented that firms with politically connected boards seem to get preferential access to credit and better treatment by the government [15,16]. These links between politics and business seem quite widespread [56] and seem to add considerable value to firms [57]. This phenomenon is more prevalent where institutions are weaker and governments more powerful but less accountable. For instance, Faccio et al. [58] find that politically connected firms are significantly more likely to be bailed out when the International Monetary Fund or the World Bank provides financial assistance to the firm's home government.
The aforementioned evidence shows that business networks, through DI, can play a valuable role for companies in dealing with an environment of constraints. However, its effects not only circumscribe on improving access to sources of financing; there is also evidence that the presence of DI improves the performance of companies. For instance, Pombo and Gutiérrez [59] find a positive relationship between the degree of board interlocks and the return on assets of privately held family business groups in Colombia. As the literature states, one of the main tasks assigned to boards is to monitor management, this role being tied to the structure of the board. Black and Kim [60] report that, in Korea, board structure affects firms' market value. They find that those firms with a minimal number of outside directors coupled with an audit committee have good corporate governance, which, in an emerging market context, positively affects market value.
In consideration of the above, this study attempts to make important contributions in the context of the influence of directorate interlocking on ESGD in emerging markets. Chile is an excellent case to study firms' DI-ESGD relation for three reasons. First, it is a developing country recently accepted into the OECD, which ensures that certain standards regarding how firms, directorates, and shareholders may organize themselves are in place. These regulations are less strict than in the US or European markets, but they comply with the OECD minimum standards. Thus, DI exists and, due to concentration of property, captures the controlling shareholder's preferences towards ESGD. Second, it is a small and open economy, where it is common to find individuals that are directors on several boards, who simultaneously act as important shareholders in other firms. Finally, Chile's cultural and economic structure permits the effect of the intensity of DI on ESGD to be interpreted as the preferences of controlling shareholders towards ESGD. Therefore, DI provides an information highway through which the relatively homogeneous values of the controlling shareholders influence the firms' ESGD decision.

The Role of Institutional Investors in the Latin American Context
During the last few decades, emerging markets have witnessed the rise of institutional investors. According to the OECD [61], institutional investors have been transformed into the most influential and largest non-controlling shareholders of listed firms in Latin America. Moreover, their role, especially after the global financial crisis of 2008-9, is crucial in the region to support the development of proper and efficient operation of financial markets, and the implementation of sound governance practices. The important role of institutional investors stems from the concentrated ownership structures of Latin American firms, where controlling shareholders can affect the allocation of resources to their benefit, at the expense of non-controlling shareholders and other stakeholders. Thus, these professional investors contribute to the development of capital markets by stimulating efficient transactions, sound risk evaluation, and a better corporate governance system [62]. Moreover, the relevance of institutional investors is reflected in their AUM, which reached USD 1500 billion at the end of 2006, growing at an annual average rate of 16% from 1999 to 2006.
In Latin American markets and, by extension, in Chile, the participation of institutional investors in the property of listed firms emerged in a particular context. First, for being a country with a legal system based on Civil-Law, and as the literature shows, because listed firms are characterized as having a high ownership concentration. The latter results in the existence of conglomerates with a great capacity for influence and negotiation [30]. As Majluf et al. [17] find, Chilean listed firms exhibit a large ownership concentration compared to companies in developed markets. Additionally, they identify the existence of economic groups formed by family-owned groups, management buyout groups, and management buyout groups coupled with institutional investors' participation. One aspect that they highlight is that shareholders who control economic groups place high value on keeping tight control of their companies. This management style or ethos generates a low turnover of shares in the stock market and crucially high interlocking directorates among Chilean boards.
Second, similar to the rest of Latin America, there are structures peculiar to corporate governance in the Chilean market. Le Fort and Walker [47] and Buchuk et al. [46] provide evidence of the widespread use of pyramid structures among Chilean firms. In particular, they state that the predominant forms of corporate structure are conglomerates or economic groups and that the most common way of separating control of firms from cash-flow rights is through tree-like ownership structures. Likewise, they state that controllers of Chilean conglomerates rely on a relatively small number of people to manage their business, and these executives mostly participate as board members of firms affiliated with their group. Consequently, directorate interlocking (DI) emerges as a distinctive characteristic of such structures of Chilean corporate governance (the latter evidence shows that Chilean corporate governance structures have certain particularities, the most prominent being that the corporate sector is highly concentrated, and that control over much of the corporate sector is exercised by a very small number of powerful groups. As a result, the stock market is rendered more illiquid than it would be otherwise, and its usefulness as a mechanism to raise investment funds is reduced [48]. Although we focus on economic and financial factors that may explain the behavior and interests of controlling groups in Chilean firms, it is important to note that idiosyncratic, historical, and social factors are also present. As Majluf et al. [17] indicate, the legal environment, tax incentives, and strategic partnerships are significant, but local historical patterns and cultural traits deeply ingrained in the Chilean management and entrepreneurial culture are also relevant).
Over the last 30 years, pension funds, insurance companies, and general fund managers have become the most important institutional investors in the country, with pension funds being the most relevant in terms of AUM and trajectory. At the end of 2017, pension managers ran about 85% of the Chilean GDP, equivalent to USD 210 billion [62]. As the literature indicates, institutional investors have the potential to play a key role in disciplining firms and limiting the extraction of rents by controllers via higher transparency in firms' corporate governance, which would thereby improve the stock market's liquidity [29,53,55].
In this market context, in recent years, responsible investment practices among institutional investors have gained notable momentum. First, in 2015, the Santiago Stock Exchange (SSE), in conjunction with S&P Dow Jones Indices, launched the Dow Jones Sustainability Chile Index, being the first local and Latin American stock index in terms of sustainability. This index seeks to measure the performance of the most sustainable companies in Chile (for more details, see https://www.spglobal.com/spdji/en/indices/esg/ dow-jones-sustainability-chile-index/#overview (accessed on 20 February 2021)). Second, since 2019, the SSE has annually conducted a survey among the most relevant local and international institutional investors in the country in order to determine their responsible investment practices in terms of knowledge, importance, and consideration of ESG variables that these investors apply in both management and investment decision-making (the 2020 survey shows that 52% of the institutional investors surveyed have an ESG policy to guide investment decisions and 37% of them are working to implement it. Regarding the capture of ESG information in Chilean companies, 74% think that the regulator should promote the standardization and higher quality of the information provided by firms, and 71% believe that a minimum of comparable ESG information should be required to be disclosed regularly. For more details, see www.bolsadesantiago.com (accessed on 20 February 2021)).
To summarize, and in order to complement the aforementioned contributions of this paper, it is worth noting that this study combines the particularities of an emerging market in terms of corporate governance, directorate interlocking, and concentration of property, among other characteristics, with the participation of experienced institutional investors in the property of firms-investors that, as global actors, are interested in improving the adoption of ESG practices among local firms. This combination of elements, coupled with the high quality and availability of data, takes the Chilean context as an optimal scenario for testing the main hypotheses of this study and to use the results of this analysis to shed light on the tension between controlling and non-controlling shareholders regarding ESGD in Latin America.

Determinants of Voluntary Disclosure
A considerable body of research studies the voluntary disclosure behavior of companies. The geographical focus of these studies is mainly developed countries and, to a lesser degree, developing countries. The general findings of these studies indicate that many of the drivers of voluntary disclosures in developed countries also apply in developing countries. A consistent finding is that board characteristics and ownership have more influence on the occurrence, quality, and extent of voluntary disclosure compared to company characteristics. For instance, analyzing the role of ownership structure characteristic on the integrated reporting policies among international listed companies, Raimo et al. [63] find a positive effect of institutional ownership on the quality of integrated reports and a negative effect of ownership concentration, managerial ownership, and state ownership. These results show how the different ownership structures influence the disclosure policies that affect corporate communications between firms and their stakeholders.
Similar results stand at the country level studying ownership and board composition. Analyzing voluntary disclosure of listed companies in China, Huafang and Jianguo [64] find that higher blockholder ownership and foreign listing/shares ownership are associated with increased disclosure. However, managerial ownership, state ownership, and legal-person ownership are not related to disclosure. Similarly, an increase in independent directors increases corporate disclosure, while CEO duality is associated with lower disclosure. They also state that larger firms have greater disclosure, while firms with growth opportunities are reluctant to disclose information voluntarily.
Barako et al. [65] analyze the effect of corporate governance, ownership structure, and firm characteristics in the behavior of listed Kenyan firms. They find that corporate governance attributes associated with voluntary disclosure are the audit committee and the level of independent non-executive directors. In addition, they find evidence of two ownership structures that influence voluntary disclosure: the proportion of foreign ownership and the percentage of stock owned by institutional shareholders. Finally, the size and the financial leverage of a company are characteristics positively related to the voluntary disclosure behavior of firms.
Finally, the boards play a key role in the voluntary disclosure behavior of firms. In the presence of agency problems, the board is one of the main control mechanisms that shareholders use to mitigate the intensity of agency costs. More importantly, if there exists a complementarity between the board and transparency, then boards foster the disclosure of information [66]. In this sense, several studies delve into how the strength of the board conditions the existence and quality of the disclosed information through which boards reduce the asymmetry of information between managers and owners [67,68]. Building on the assumption of complementarity between certain board characteristics and firm transparency, Jensen and Meckling [69], Donelly and Mulcahy [67], and Frías-Aceituno et al. [68] test the positive relation between the boards' strength and firms' disclosure and its quality. In these studies, a board's strength is given by its size, independence, activity, and diversity. In this regard, the individual "board" characteristics that positively affect voluntary disclosure are size, independence, and diversity.
In summary, voluntary disclosure by companies in both developed and developing countries depends on many factors. Elements regarding the cultural environment in which firms operate, coupled with corporate governance, ownership structure aspects, and company characteristics, are relevant for explaining the voluntary disclosure of firms. Company features such as size, listing, financial leverage, profit, and growth have an impact on disclosure. Moreover, corporate governance aspects, such as board characteristics and composition and ownership structure, also influence the voluntary disclosure of companies. Although this phenomenon has been studied extensively in developed countries, the peculiarities of developing countries mean that a higher level of knowledge of this phenomenon is still required in emerging markets. ESG disclosure in Latin America is rarely studied. This scarcity of empirical evidence, in conjunction with the unique context where its companies operate, may result in recommendations that could be counter-productive in terms of ESG disclosure of Latin American companies when extrapolating previous results from developed countries [5].

Hypotheses
Controlling for the determinants of stakeholders' influence on the firms' ESGD in a Latin American context [8], the previous discussion implies two hypotheses that jointly allow us to study the tension between the preferences of the controlling shareholders and institutional investor shareholders regarding ESGD. Our first hypothesis is the following.

Hypothesis 1 (H1). The intensity of directorate interlocking of a firm is negatively correlated with ESGD.
We expect the intensity of the DI of a firm to be negatively correlated with ESGD because DI's intensity captures the influence of controlling shareholders, which we assume are current value maximizers. DI intensity reflects the position of a firm in the DI network, which is the outcome of the governance practices of the conglomerates that participate on the firms' boards. Our second hypothesis is the following.

Hypothesis 2 (H2). The percentage of participation of institutional investors in a firm is positively correlated with ESGD.
We expect the participation of institutional investors to be significantly correlated with ESGD because of their growing interest in ESG investing and their growing participation in the property of Chilean firms, which grants them a larger representation and influence on boards' decisions.

Data and Sample
The sample of firms that we analyzed for testing our hypotheses corresponds to the 124 firms included in the General Price Index IGPA of the Santiago Stock Exchange (SSE) for the period 2015-2017. This dataset is free from survivorship bias since we considered all the firms that were included in the index during that period. The IGPA index groups together a large part of the shares traded on the SSE; therefore, it accurately represents the main market trends. IGPA is the oldest index on the market, dating back to 1954, and, since then, it has constituted one of the main market indicators for the Chilean stock market. The market capitalization of the component companies of the IGPA is equivalent to MMUSD 208,930, which, at the end of 2016, represented 89.65% of the market capitalization of all listed firms in the SSE.
The data that we used to build the variables that test our main hypotheses belong to two sources: one related to financial information of the companies listed in the SSE, and the other linked to the information of the board members of the firms. For each firm, we obtained the financial data and the ESG indicators from the Bloomberg Terminal (www.bloomberg.com (accessed on 20 February 2021)). In the same way, we collected the data on each firm to build indicators of the stock's activity from the website of the SSE. We gathered the data regarding the boards' members of the sample firms from the Chilean Commission of Financial Markets (CMF, for its acronym in Spanish, www.cmfchile.cl (accessed on 20 February 2021)) by web scraping. The information in the CMF identifies yearly each director by their full name and national ID number and each firm by its full name and its national firm ID number. In this way, it is possible to identify and link each director with the firm or group of firms that they belong to (we collect data made public by the Commission for Financial Market (CMF, www.cmfchile.cl (accessed on 20 February 2021); it is similar to the US Securities and Exchange Commission) to construct the property structure of main stockholders and to identify all the participants on the board of directors for the firms regulated by them. This dataset uses a unique national ID (consistent in various databases) to record the participation of individuals on boards and firm ownership. Therefore, this ID number allows us not only to construct the DI networks but also to establish with certainty specific details of directors' involvement (e.g., if a specific individual listed as a director for firm A is also the owner of participation on firm B)).

Model
To the extent that the interest of controlling and non-controlling shareholders groups differs in ESGD, the firms' decision on the adoption of ESG reporting is the outcome of how the eventual tension between these two sets of interest is resolved within the firms' boards (Cossin [70] identifies four tiers of conflicts of interests within boards. One of these layers arises when some members of the boards tend to act in their own interests or their group's at the expense of society's interest). As mentioned, the Chilean context provides an ideal environment to study this possible tension between the two groups because of the relevance that ESG investing has acquired in recent years. This investing force has driven institutional investors to value the broad effects of good ESG performance more than the merely private impact on operational or investment costs required by an ESG management and reporting system [2]. The latter does not occur with the same intensity for the case of the "traditional" controlling Chilean shareholders [17].
To testing our hypotheses and capture the eventual tension between the preferences represented by the controlling shareholders' directors and the institutional investors' directors, we consider the following Logit econometric model: Our parameters of interest are the γ coefficients. The parameter γ 1 captures the influence of controlling shareholders on ESGD through the intensity of directorate interlocking (DI, variable IL it ). We use the IL it as a proxy of the ease with which the values and preference of controlling shareholders on ESGD of other firms reach firm i at time t through board members. Thus, the intensity of DI captures the ease with which the ethos of controlling shareholders reaches the firms' boards. On the other hand, γ 2 captures the influence of institutional investors (variable Teninst it ). We directly measure the influence of institutional investors by considering their participation in the property of firm i during the year t. This measure captures the aggregated influence and participation of institutional investors acting as non-controlling shareholders on ESGD.
We measure the decision regarding the adoption of ESG reporting with the dummy variable (ESGD it ) that takes value "1" if the firm i reports non-financial information pertaining environmental, social, or governance issues as reported in Bloomberg on year t, and value "0" otherwise (for a broader review of the literature to explain the selection of regressors of this first set of determinants, see [8]). Finally, we control for several variables that, according to [8], influence the probability of ESG disclosure for Latin American firms (we depart from the specification of Duran and Rodrigo [8] in two aspects. First, we do not consider the quality of the regulation because all firms included in our study are subject to Chilean regulations. Second, we expand on Duran and Rodrigo's rationale to include firms' "beta" beyond risk. Based on their literature review, they argue that riskier firms (with larger "beta") use ESG reporting to hedge against higher risk levels because ESG reporting provides more information, which reduces uncertainty. Thus, ESG reporting becomes a signal of good public image, which compensates for other (financial or economic) risks. Analogously, we argue that more liquid firms also have an incentive to provide more information than less liquid firms. The latter occurs because more liquid firms are more exposed by the visibility that high liquidity levels provoke. Thus, we use a firm's liquidity as a proxy for the incentive to signal a good public image to the market). These control variables are represented by the determinants with β coefficients. Ln asset it represents the natural logarithm of the total assets of firm i during year t. Mg ebitda it corresponds to the EBITDA (earnings before interest, taxes, depreciation, and amortization) margin of firm i during year t. Rotact it measures the asset turnover of firm i during year t. Ln leverage it measures the natural logarithm of the financial leverage of firm i during year t. Dum degint it is a dummy variable that takes the value 1 if firm i exports during year t, and 0 otherwise. Price to book it measures the price to book ratio of firm i during year t. Liqbur it equals turnover to market capitalization and represents the stock liquidity of firm i during year t. Finally, Teninst it corresponds to the percentage of institutional investors' participation in the property of firm i during year t. Table 1 contains descriptive statistics for the dependent and independent variables. Regarding controlling groups and their behavior, recall that we posit that directorate interlocking (DI) provides an information highway through which management information, such as companies' practices, propagate from one firm to another. The intensity of DI can be measured in different forms. In this research, we apply social network methodologies because its framework allows the interaction structure of a complex system to be represented, such as directorate networks and their relationships with firms.
Previous work uses social network analysis to study how the position of a firm in the DI network relates to ESG performance. Harjoto and Wang [71] use social network analysis to capture the social capital that firms use to improve their ESG performance. Although we also use social network analysis to capture the position of firms in the DI network, the interpretation that we give to the centrality measures is not of social capital as our endogenous variable is different-ESG disclosure, not performance, as in Harjoto and Wang [71]. In our setting, a larger centrality (closeness, betweenness, etc.) implies a higher level of exposure to the values and objectives of controlling economic groups, which gives rise to DI as they appoint more directors in different firms. Accordingly, we capture the intensity of interlocking by applying different measures of interconnectedness. A simple measure of connectedness is having more or fewer connections (links) in the interlocking network. This measure is known as the degree centrality of a firm (ln_deg). Thus, in the context of our study, the latter measure of interlocking intensity is equivalent to sharing at least one director with more or fewer different firms. An alternative measure of connectedness relates to how many links on average separate one firm from another in the network. This measure is known as closeness centrality (ln_clo). Similarly, one could also consider that firm i is more connected if shorter paths that connect any other two firms k and l pass through i. In this case, a more connected firm is one that plays the role of a bridge more often. This intuition is captured by the concept of betweenness centrality (ln_bet). Finally, another interpretation of the interlocking intensity of a firm could be related to the number of links that firms' neighbors have. The latter interpretation could be regarded as the prestige of a firm in the interlocking network, and we measure it with the eigenvector centrality (ln_eig) of a firm in the network.
In our analysis, we study the intensity of DI considering each of these individual measures one by one. Because each interlocking measure has nuances in its interpretation, we also perform a likelihood-ratio test based on comparing unrestricted and restricted models with all four measures to study whether interlocking affects ESGD. All the econometric models were performed by applying clustered standard errors.

Empirical Strategy
To study the aforementioned hypotheses, we analyze our model in two stages. First, we study the significance and sign of the γ parameters. Then, we apply nested model tests to study the influence of each channel of our model in ESGD. The idea behind this approach is to determine whether both channels are relevant for explaining the probability of reporting ESG indicators. In this sense, it is interesting to compare the strength of the controlling groups and the institutional investors on the ESG behavior of firms. We think that this analysis will be of interest to other markets and countries that will confront or are facing characteristics similar to Chilean markets in terms of financial development, corporate governance issues, and ESG contexts. As mentioned, in each stage, we control for variables that positively correlate to ESGD in the Latin American context according to Duran and Rodrigo [8].

General Results
As indicated in Sections 2 and 3, we analyze H1, H2, and, finally, both hypotheses in conjunction. First, when we study the relation between the intensity of DI on the ESGD, we can observe in Table 2 that there is an inverse DI-ESGD relationship. Each of the measures that we use to capture the intensity of DI (degree, closeness, betweenness, and eigenvector centrality) is negatively correlated with probability reporting ESG performance (ESGD). These results suggest that the implicit influence channel that directorate interlocking provides between firms in Chile is a relevant factor for explaining the ESG reporting behavior of listed firms, and that controlling shareholders are not prone to ESGD. In other words, when a firm is more intensively interlocked with other firms due to sharing board members, the probability of reporting ESG indicators is lower. The evidence provided by Table 2 suggests that through DI, an avoidance sentiment towards ESGD spreads through the DI network.  Second, when we study the effect of institutional investors' participation on the probability of ESG reporting by Chilean listed firms, we find that there is partial evidence of a positive relationship between institutional ownership and the likelihood of ESG reporting. As Table 3 shows, when including industry-fixed effects, the statistical significance of the variable ten_inst disappears, whereas it continues to be significant when we include only year-fixed effects. Together, the results of Tables 2 and 3 suggest the existence of a weak tension between the preferences of controlling and non-controlling shareholder towards ESGD. This lack of consensus arises from the positive (negative) influence that institutional participation and the controlling groups have on the probability of ESGD. On one hand, the former group favors ESGD, and on the other, the latter group does not.
In the next step, we compare collectively the relevance of both DI and institutional participation on the probability of ESG reporting by the Chilean listed firms. The evidence indicates that the signs of the previous results remain. Nevertheless, the significance of the variables changes when applying fixed effects by industry, year, and industry-year. As Table 4 shows, after controlling for industry-and year-fixed effects, the inverse relationship of interlocking and the probability of ESG reporting is supported, while, for the case of institutional participation, the previously positive effect disappears. The latter indicates that when the influences of the preferences of controlling and non-controlling shareholders are considered simultaneously (main model presented in Section 3.2) through the intensity of DI and institutional investors' participation, respectively, only the intensity of DI has a significant correlation with the probability of reporting ESG disclosure performance (ESGD).
Finally, to further study the strength of the tension between the preferences of the controlling and non-controlling shareholders on ESGD, we apply nested model tests based on the models presented in Table 5. The latter analysis confirms that the institutional investors' participation variable is not significant to explain the probability of reporting ESG variables. As shown, the LR-Chi2 and prob >Chi2 indicators for each of the models applied with industry-, time-, and industry-time-fixed effects point out that the institutional ownership dimension captured by the ten_inst variable would only be statistically significant with p-values higher than 29.08%. That is, with a significance level of 10%, the null hypothesis that the econometric model that excludes the ten_inst variable provides a better estimate of the probability of reporting ESG indicators that cannot be rejected. Thus, the evidence suggests that the main influence on ESGD is exerted by controlling shareholders who are only weakly opposed by the non-controlling shareholders. Hence, there is evidence of only a weak tension between these two different preferences towards ESGD in the Chilean boards, with the controlling group's preferences being those that play a determining role in the ESGD decisions of the firms considered in this study (publicly traded firms in the SSE).
In summary, our main findings provide evidence that support Hypothesis 1. Therefore, directorate interlocking (DI) phenomena are a significant factor to explain the probability of ESG reporting among the Chilean listed firms. In addition, the sign of the effect is negative, suggesting a negative relationship between interlocking directorates of Chilean listed firms and the probability of reporting ESG variables. However, Hypothesis 2 is not supported by the evidence, i.e., institutional investors' participation is not correlated with ESGD (columns 2, 4, 6, 8, 10, 12, 14, and 16, Table 5).

Discussion
The evidence presented in our analysis suggests that, at least in the case of ESG performance disclosure, directorate interlocking generates an influence channel between firms. Directors spread values, experiences, and know-how (information) from one firm to another. In our model, the network of firms created by DI is an information highway where we interpret that the intensity of DI is the exposure to the values, experiences, and know-how, external to the firm, that influence the boards' decisions to disclose ESG performance (H1). The relevance of the latter exposure, in the context of emerging markets, is that because emerging markets are concentrated, it is very likely that the more exposure to a firm's information, the greater the influence of the ethos of other firms on strategic decisions such as ESG reporting. The evidence provided by our analysis in the Chilean case is consistent with the findings of Majluf et al. [17], Agosin and Pastén [48], and Le Fort and Urzúa [72]. That is, firms with greater exposure to information, i.e., intensity of DI, are less prone to reporting ESG performance because the controlling shareholders' preferences as expressed by the board members mean that they elect place relatively less weight on ESG performance reporting than on other strategic corporate aspects.    In order to delve more deeply into the mechanism through which information from other firms, unlike a given firm i, affects the ESG reporting decision of firm i, we considered different measures of interlocking intensity. Through firm i's degree centrality, we measured exposure as the portion of firms directly connected to firm i; firm i's closeness centrality measures exposure as the average distance (in links) to other firms in the DI network; firm i's betweenness centrality measures exposure as the proportion of the shortest path between how firms l and k pass through firm i; and eigenvector centrality measures exposure to information as how central (in the degree sense) the neighboring firms of i are.
All of the DI intensity measures that we consider are significant individually. This finding is consistent with the concentration of property and it implies the homogeneity of preferences regarding ESGD. The larger the concentration of property, the more difficult it is to escape the influence of information external to the firm (values, experience, and know-how) regarding ESGD. Thus, the channels of exposure (number of directly connected firms, the average distance to other firms, the "bridge" role with respect other firms, and the exposure to strong influence) through which DI affects ESGD appear to all be relevant.
The negative and significant coefficients of the centrality measures in Tables 2, 4 and 5 suggest that the influence of economic groups in the Latin American context reflects the "current value-maximizing behavior" [17,73] of these groups. Especially in the Chilean case, the latter reflects the history of the leaders of those economic groups, who initially played the role of managers for many years and now are at the top levels of businesses [17]. We argue that this role causes controlling shareholders objectives to differ from the interests of minority shareholders, represented by institutional investors, who are more in tune with a broader scope of stakeholders that focus on the long-term value of the firm. This relationship is consistent with the underlying agency problem in this tension between directors appointed by controlling shareholders and minority shareholders in Latin America. Thus, we conjecture that the negative coefficient of the centrality measures would reflect a higher level of external influence to the firm caused by the director appointments that produce DI in the local context [16,32,53,54]. In this sense, our results are in line with the findings of Husted and de Sousa-Filho [5] and Correa-Garcia et al. [74].

Conclusions
The main contribution of our paper is to explore how the interests of controlling economic groups and the interests of institutional investors influence ESG disclosure in the Latin American context. Thereby, we analyze the eventual tension between groups of board members and ESG disclosure. This tension regarding ESG disclosure emerges because, on Latin American boards, coexisting directors are appointed by controlling shareholders that often designate the same director in several firms, and directors that are appointed by minority shareholders such as institutional investors. To the best of our knowledge, these questions have not yet been addressed in the Latin American context. Our main findings indicate that the intensity of directorate interlocking is negatively correlated with the probability of disclosing ESG performance among the Chilean listed firms. On the other hand, our results suggest a weak relation between the participation of institutional investors and ESGD. These correlations are robust after controlling with industry-and timefixed effects. Thus, although we find some evidence of tension within the boards regarding ESG disclosure, it is the position of the controlling shareholders through their elected directors that drives each company's decision on ESG disclosure. We further conjecture that this weak tension among the interests of controlling and non-controlling shareholders within the board members evidences an emerging, and eventually growing, dissensus within Latin American boards regarding the disclosure of ESG performance.
An interesting aspect of the latter findings is that the intensity of directorate interlocking (degree, closeness, betweenness, and eigenvector) and the probability of ESG performance disclosure have an inverse relationship. The evidence from the Chilean case suggests that the negative association between the intensity of DI and ESGD occurs because ESG performance disclosure is not yet a major strategic priority of controlling shareholders.
Our literature review suggests that the latter would be caused by the concentration of property in emerging markets where controlling economic groups have a focus on maximizing the current value of the firm.
Another relevant aspect of our findings is the absence of consistent evidence regarding the effects of the participation of institutional investors on the probability of disclosing ESG performance. Although institutional investors have growing participation in publicly traded firms in emerging markets, their presence on the boards of directors has not yet gained sufficient strength to have a clear influence on ESG performance disclosure in the emerging economy context. As in other studies, we find that directors that represent minority shareholders promote ESG disclosure and, to some extent, greater transparency. Therefore, in the current global context of transparency and investment discipline, the relevance of institutional investors, acting as delegated portfolio managers, increases as active ESGD promoters among the Latin American financial community.
Our study has some limitations. First, we cannot directly generalize our conclusions to the rest of the countries in the region. Although the Chilean capital market is similar enough to other important markets in the region, the singularities of the Chilean firms listed on the stock exchange lead us to suggest that these results should be taken with caution. Among the similarities between Chile and Latin America in general that provide the basis to use Chile as a useful case study for Latin America are the low liquidity and high concentration of firm ownership in the financial market. Nevertheless, the particular historical evolution of the Chilean economy during the last 30 years motivates caution in generalizing our findings to Latin America as a whole and emerging markets in general. Second, the companies in the sample correspond to the main firms included in the broadest index of the Chilean stock market; therefore, we cannot necessarily extend these results to the rest of the locally listed firms that do not belong to the IGPA index.
Our study has practical and theoretical implications. First, we evaluate two specific hypotheses that are pertinent to the reality of Latin American boards, namely the influence of institutional investors and the influence of controlling economic groups through the interlocking of directors. Second, from a practical point of view, our research provides information to firms and their boards regarding the factors that affect the probability of reporting on ESG performance indicators, and the possible conflicts that would arise between different groups of shareholders if there is no consensus regarding the ESG phenomenon within boards. Third, from the perspective of institutional investors currently interested in incorporating ESG indicators into their investment models, our results emphasize the need to push improvements in the ESG reporting of possible investable assets such as stocks and bonds. Moreover, our analysis and our findings also highlight the relevance of promoting ESG performance disclosure by Latin American firms as a means to reduce the concentration of property by facilitating strategic and financial analysis to non-controlling shareholders. In the context of financial constraints, as is the case of emerging economies, the latter could improve the access to funds to publicly traded firms, thereby reducing financial costs and generating value to them. Finally, our results support regulators in the task of advancing the regulation of ESG practices in the region. As our results indicate, the active participation of all stakeholders is essential, especially the support of the controlling shareholders.
Although our work sheds light on the probability of ESG reporting, we believe that advancements are necessary regarding the following aspects. First, extending the longitudinal study to a greater number of years and firms in the region would provide the possibility of a greater generalization of the results. Second, it would be interesting to explore the behavior of firms belonging to industries other than those considered in this study in order to delve into possible differences. Third, analyzing different levels in the ESG information report by environmental, social, and corporate governance dimensions could signal the relative importance that firms give to these activities.
Finally, we believe that some questions require further attention. First, it is worth asking whether the results of this research would be similar in developed countries where interlocking, the participation of institutional investors, and the regulation on them are different from those observed in Chile. Second, it would be interesting to analyze whether the dimensions that we explored in this research are also significant in developed contexts. We believe that these questions are relevant to the development of ESG practices and to improving our understanding of the dissemination mechanisms of ESG practices in both developed markets and the Latin American markets.
Author Contributions: Both authors contributed equally to the development of this article, specifically in its conceptualization, conceptual framing and hypothesis construction, methodology design, data analysis, and writing-review and editing. Both authors have read and agreed to the published version of the manuscript.

Data Availability Statement:
No new data were created or analyzed in this study. Data sharing is not applicable to this article.